Michael Slavik

Stochastic Broadcast for VANET

Vehicular Ad-hoc NETworks (VANETs) are an emerging important type of wireless ad-hoc network. Unlike many other types of MANETs, VANET applications such as traffic data dissemination are inherently broadcast oriented and require communication protocols to be anonymous and scalable. Simple broadcast flooding satisfies these requirements, but its performance is highly dependent on network density and may lead to the broadcast storm problem. This work is the first we are aware of to propose stochastic broadcast as a solution for VANET. Stochastic broadcast instructs nodes to rebroadcast messages according to a retransmit probability. Such a scheme is private since node identification is unnecessary, however it has an undesirable dependency on vehicle density in the same manner as simple flooding. To solve this problem, we demonstrate the link between the mathematical science of continuum percolation and stochastic broadcast. Specifically, that the critical percolation threshold in continuum percolation (≈ 4.5 expected neighbors) translates to the wireless broadcast context. Then we show that nodes can tune the performance of the broadcast system to efficient levels by adjusting the retransmit probability so the apparent density of the network approaches the critical threshold.

Spatial Distribution and Channel Quality Adaptive Protocol for Multihop Wireless Broadcast Routing in VANET

Multihop wireless broadcast is an important component in vehicular networks. Many applications are built on broadcast communications, so efficient routing methods are critical for their success. Here, we develop the Distribution-Adaptive Distance with Channel Quality (DADCQ) protocol to address this need and show that it performs well compared to several existing multihop broadcast proposals. The DADCQ protocol utilizes the distance method to select forwarding nodes. The performance of this method depends heavily on the value of the decision threshold, but it is difficult to choose a value that results in good performance across all scenarios. Node density, spatial distribution pattern, and wireless channel quality all affect the optimal value. Broadcast protocols tailored to vehicular networking must be adaptive to variation in these factors. In this work, we address this design challenge by creating a decision threshold function that is simultaneously adaptive to the number of neighbors, the node clustering factor, and the Rician fading parameter. To calculate the clustering factor, we propose using the quadrat method of spatial analysis. The resulting DADCQ protocol is then verified with JiST/SWANS and shown to achieve high reachability and low bandwidth consumption in urban and highway scenarios with varying node density and fading intensity.

Statistical Broadcast Protocol Design for Unreliable Channels in Wireless Ad-Hoc Networks

Broadcast is a critical component in ad-hoc wireless networks. This paper examines the effects that channel unreliability have on the performance of broadcasting protocols. We show that the distance method of statistical broadcast performs poorly under adverse channel conditions. We then describe how to design a version of the distance method that is tolerant of transmission errors by adjusting the statistical variable threshold according to the channel conditions. The resulting protocol is then compared with the Double Covered Broadcast (DCB), a topological protocol designed to handle similar circumstances. The modified distance method protocol is shown to give similar reachability characteristics to DCB while using far less transmissions.

Applying machine learning to the design of multi-hop broadcast protocols for VANET

Multi-hop broadcast is an important component in ad-hoc wireless networks. Some vehicular network (VANET) applications in particular use broadcast communications extensively. We propose using the distance-to-mean method to facilitate these applications. The performance of this method depends heavily on the value of the decision threshold, and it is difficult to choose a value for that threshold that results in good performance across a wide range of network scenarios. Node density, spatial distribution pattern, and wireless channel conditions all affect the optimal statistical threshold value. VANETs exhibit wide ranges of these factors, so protocols designed to support these applications must be adaptive to those variations. In this work we address this design challenge by using black-box optimization algorithms based on machine learning techniques such as genetic algorithms and particle swarm optimization to automatically discover a decision threshold value for the distance-to-mean method that is simultaneously adaptive to node density, node distribution pattern, and channel quality. The resulting decision surface is a function of the number of neighbors N, the quadrat statistic Q, and the Rician fading parameter K. The result is the Statistical Location-Assisted Broadcast (SLAB) protocol. Evaluations using JiST/SWANS show SLAB achieves high reachability and efficient bandwidth consumption in both urban and highway scenarios with varying node density.

Adapting statistical broadcast to linearly oriented networks for VANETs

Broadcast is a critical component in ad-hoc wireless networks. Some vehicular network (VANET) applications in particular use broadcast communications extensively. VANETs exhibit a wide variety of node density and distribution patterns, so broadcast protocols designed to support these applications must be adaptive to those conditions. We show that the distance method of statistical broadcast can be used to design a protocol that performs well in one-dimensional or two-dimensional uniformly distributed networks, but not both. We propose using the quadrat method of spatial analysis to characterize the distribution pattern at each node and use the resulting metric, K, as a factor in computing the statistical threshold function. The result, the Distribution-Adaptive Distance (DAD) method, is shown to exhibit a high level of reachability and efficient bandwidth utilization across a range of distribution patterns from purely linear to purely two-dimensional and sparsely distributed to densely distributed. More generally, the design methodology presented in this work provides a procedure enabling statistical broadcast protocol designers to build protocols that are adaptive to both node density and node distribution. This capability is a key prerequisite for design of practical broadcast protocols to support VANET applications.

Statistical broadcast protocol design with WiBDAT: Wireless Broadcast design and analysis tool

Multi-hop broadcast routing is a critical component in ad-hoc wireless networks. Vehicular ad-hoc network (VANET) applications for example use broadcast communications extensively. Efficient broadcast protocols are required to enable these applications. Multi-hop broadcast algorithms can roughly be classified as either topological or statistical. Statistical protocols are well-suited to applications such as VANET, because they are tolerant of rapid changes in network topology and can be designed to accommodate transmission failures due to fading and collisions. However, unlike topological protocols, statistical methods are highly stochastic in nature so it is not possible to analytically prove these algorithms will connect all nodes in the network. Instead, designers attempt to measure and minimize the probability nodes will not receive a broadcast message. Thus simulation plays an important role in the design and validation of these protocols. The key design factor in statistical protocols is the rebroadcast decision threshold curve. There are no analytical tools for creating these threshold curves, so like many stochastic systems, simulation models are required to develop them. Empirically discovering good threshold curves using simulation models requires many simulation iterations, and thus can be extremely time consuming if these iterations are slow. As a solution, this work introduces the Wireless Broadcast Design and Analysis Tool (WiBDAT), a fast and scalable high-level network simulator built from the ground up specifically for wireless broadcast. WiBDAT provides tools for quickly designing efficient statistical broadcasting protocols and evaluating their performance at a high level. WiBDAT is easy to use and extend, implements a wide variety of broadcast protocols, and provides clear results including innovative visualizations of node and transmit densities.

The distance-to-mean broadcast method for vehicular wireless communication systems

Broadcast is a critical component in embedded communications systems. Some vehicular network (VANET) applications in particular use broadcast communications extensively. Statistical broadcast methods offer an efficient means of propagating broadcast messages in this context due to their low overhead and high efficiency.

Analytical model of energy consumption in hierarchical wireless sensor networks

Here we propose an analytical model of energy usage hierarchical wireless sensor network protocols such as LEACH [1]. LEACH has been used as a foundation for much of the wireless sensor network research, so results addressing the underlying nature of these systems build understanding of a wide range of existing protocols. First, we derive the probability density functions of the distance to the nearest cluster head, the distance to the nearest cluster head squared, and the expected number of nodes per cluster. These results are then applied to produce a model of the amount of average energy used per node for each data report. Using this model we give an analytic expression for the optimal cluster-head selection rate. The model is further improved to give an approximate adjustment for distortions in energy usage near the edges of the field. This final model is then compared to the simulation results presented in the original LEACH paper [1] and shown to be accurate. Sensor network protocol researchers can apply results presented here to improve performance of existing systems by making the cluster head selection rate adaptive to external factors.

Designing statistical multi-hop wireless broadcast protocols using confidence levels from stochastic models of reachability

Multi-hop wireless broadcast is a critical component in ad-hoc wireless networks. Vehicular Ad-hoc Networks (VANET) in particular utilize broadcast as a primary communication mechanism in many applications. Because of their simplicity, statistical broadcasting methods have been adopted in several VANET broadcasting schemes. However, it is difficult to develop analytical models of these protocols, therefore no results are known proving reachability characteristics for them like there are for topological protocols such as Multi-Point Relaying (MPR).

A hybrid data dissemination scheme for VANETs

Data dissemination in Vehicular Ad hoc NETworks (VANETs) is a key research area with applications that range from traffic flow information to weather reports and related information. Delivering this data with maximum efficiency to a wide area is the focus of research in data dissemination protocols. Data propagation methods preferred for these specialized ad hoc networks fall under two mutually exclusive categories namely store-and-forward and multi-hop broadcasts. Neither of these two categories is noticeably superior and each has its strengths and weaknesses which we outline in this paper. We propose a novel hybrid data dissemination scheme that capitalizes on the strengths of both these categories and mitigates the weaknesses by combining store-and-forward and multi-hop broadcasts. We simulate the proposed hybrid scheme on the JiST/SWANS simulator and evaluate its behavior and performance against pure store-and-forward and pure multi-hop broadcasts by varying the aggregation ratio, number of nodes, and broadcast zone radius.