Retransmission and backoff strategies for wireless broadcasting☆
Introduction
One way of performing network-wide broadcasting is by “flooding” the network with the broadcast message. Flooding is carried out by having each node retransmit the broadcasted message after receiving it for the first time.
Flooding is important because it is the basis for performing route discovery in mobile ad hoc networks (MANETs). Link-state routing protocols also rely on flooding for distribution of link-state information. There are other broadcasting techniques that send out-of-band messages to build distribution trees. However, schemes that do not require out-of-band messages continue to be the best general-purpose solution for path discovery and link-state routing as their performance does rely on assumptions about mobility or broadcast frequency.1
Flooding is an unreliable operation with no acknowledgment mechanism in place; this is not a major concern for path discovery or link-state routing as 100% reliability is not required. Flooding is able to distribute broadcast messages to as many nodes as possible using very little effort. Analysis of some of the broadcasting strategies presented in this article show that under reasonable assumptions that reliability can be at least as good as in flooding. If such analysis is not possible for a particular case, then simulations will be used to show that their reliability is comparable to that of flooding.
This article only considers broadcasting algorithms that do not require out-of-band transmissions such as hello messages. The cost of transmitting “hello” messages often cannot be justified. Thus, out-of-band overhead narrows the application scope and make it challenging to conclusive compare other broadcasting algorithms that do not use out-of-band messages.
Flooding generates more overhead than necessary because, depending on the node density, many or most retransmissions are redundant. A retransmission is said to be redundant if all the neighbors of the transmitting node have already received the message. If a transmission is non-redundant then its additional coverage is not null2, and in this case the additional coverage of a retransmission is the percentage of the transmission range that has not been covered by neighbor nodes.
Redundancy can be reduced (respectively eliminated) when retransmitting occurs only if the additional coverage of the retransmission is large enough (respectively non-zero) to warrant the additional overhead. This can be achieved if each node knows exactly the locations of many of the other nodes – e.g. its neighbors, but in a dynamic network it is quite beneficial to avoid maintaining this information. Instead, each node u receiving the first duplicate of the broadcast message would postpone its retransmission for a short backoff time, and would drop this message completely if within this backoff time other retransmission made by other nodes made u’s retransmission redundant. Otherwise u would retransmit once the backoff time has elapses. In this paper we study different method by which partial information “sensed” by u can suggest that this retransmission is redundant. For example, if the location of the transmitting neighbors is known, then the additional coverage can be deterministically computed. If the precise location is unknown, then other information such as distance or angles between neighbors or number of duplicates received can be used to estimate the expected additional coverage. Different technologies such as GPS, angle of arrival (AOA) and received signal strength (RSS) can be used to gather such information.
This hold and suppress approach to broadcasting is by no means novel. However, this article makes several contributions in the area of wireless broadcasting. Namely, we redefine a broadcasting algorithm as being composed of two strategies. The retransmission strategy refers to how it is decided if a scheduled retransmission is canceled. The backoff strategy determines the manner in which the backoff time is chosen. Most of the previous work implicitly falls only under the category of retransmission strategy. We also propose several retransmission and backoff strategies and a comparative analysis is presented between existing algorithms and the strategies proposed herein. Simulation experiments and analysis are used throughout this work to study or demonstrate the properties and performance of specific strategies or to obtain results of a more general nature. Strategies are also evaluated with respect to their impact on routing protocols that rely on broadcasting to perform path discovery. The purpose of this evaluation is to determine which strategies result in more stable routes.
By considering the distribution of nodes in specific applications it is possible to design more efficient retransmission strategies. The second part of this paper analyzes the problem of broadcasting when nodes are assumed to be arranged on a strip. Such arrangement occurs in vehicular broadcasting applications. We present the Strip Broadcasting (SB) retransmission strategy that can be modeled as a one-dimensional problem to significantly reduce the number of retransmissions. A whole new range of vehicular information services can be made possible by relaying information using vehicle-to-vehicle communications (V2VCOM). The first class of services that comes to mind is traffic alerts about different upcoming situations such as accidents, construction zones, or traffic jams.
With the exception of some cases such caravans or convoys, communication between specific vehicles is rarely an issue. Broadcasting is a more natural communication primitive for this type of environment. Accordingly, the goal is to relay information between vehicles for a certain distance or for a given number of relay hops. Omnidirectional transmissions are used because the road and the vehicles it contains can have an arbitrary direction with respect to the transmitter’s frame of reference.
Previous algorithms are suboptimal in vehicular environments because they perform considerably more retransmissions than necessary. For example, vehicle b shown in Fig. 1 has received duplicates from both a and c. Existing broadcast algorithms would require a retransmission by b because the shaded areas in its range of transmission have not been covered. However, the strip of road in the figure is entirely covered by previous retransmissions.
The rest of this article is organized as follows. Section 2 presents a summary of the related work. Section 2.1 describes the simulation environment used to obtain the experimental results presented throughout this article. Section 3 presents flooding as a combination of retransmission and backoff strategies, and provides interesting experimental results about the relationship between backoff, collisions and reliability. Section 3 proposes new retransmission strategies, and Section 3.9 provides a comparative simulation analysis. Backoff strategies are studied in Section 4. Section 5 presents an interesting study on the effect that retransmission and backoff strategies have on the performance of path discovery. Finally, Section 8 provides some concluding remarks.
Section snippets
Related work
Sze-Yao et al. [4] observed that serious redundancy, contention, and collision could exist if flooding is done blindly. Collectively, they refer to these problems as the broadcast storm problem. As a solution, they introduce several retransmission strategies, including the counter-based, distance-based, location-based, and cluster-based schemes. Williams and Camp [5] present an analysis of existing broadcasting schemes and is an excellent reference on the topic of wireless broadcasting.
Backoff strategies for flooding
Flooding is often enhanced by waiting for a short and uniformly distributed random backoff time before forwarding a message in an effort to reduce the number of collisions. As such, flooding can be expressed as a typical hold-and-suppress broadcast with a very simple retransmission strategy where messages are never suppressed. Consequently, if the distribution is uniform, this algorithm only has one possible parameter: The maximum backoff time.
The purpose of the timer at the MAC for pure
Backoff choices
Timer-based contention has been previously studied in the context of location-based routing [19], [20]. However, the backoff procedure by which nodes postpone retransmissions in network-wide broadcasting has not received much attention in the past. We introduce the concept of backoff strategy and define it as having three components:
- (1)
The backoff magnitude: How much time to wait. Usually specified as a maximum or average value.
- (2)
The backoff function: How are backoff times assigned to different
Routing
Flooding is primarily used in routing. It is the basis for performing path discovery in on-demand routing protocols [21], [22] that are commonly used in multi-hop wireless networks. This section studies the effects that different retransmission and backoff strategies may have on path discovery.
Path discovery from source s to destination d works by having s broadcast a route request (RREQ) that will eventually arrive at one or more nodes with a valid routing entry for d (possibly d itself).
Strip broadcasting. broadcasting between moving nodes
If nodes (vehicles) are assumed to be located on a strip (road) and the range of transmission is carefully chosen with respect to width and curvature of the strip, then in most situations a retransmission is deemed to be redundant once a message has been received from the front and back of the vehicle. Accordingly, we propose the Strip Broadcasting (SB) algorithm shown in Algorithm 1, where scheduled retransmissions are canceled once the message is received from both directions. Vehicles could
Simulation results
The experimental results presented in this section evaluate the performance properties of the algorithm. Such properties include retransmission overhead, coverage (reliability), scalability and latency. The SB algorithm was implemented at the application-level, meaning that the backoff period of the SB algorithm takes place outside of the MAC layer. Once the backoff period is complete, the packet travels down the protocol stack and is handed to the MAC layer, where an additional backoff period
Conclusions
In this paper, we study the application-level backoff timers necessary for effective broadcasting in a wireless ad hoc network. We observe that the application-level backoff time required to achieve 100% coverage decreases with increasing node density. At low node densities, smaller application-level backoff timers lead to collisions at the MAC layer, achieving less than 100% coverage. Thus, application-level backoff timers are critical at lower node densities. With this observation, we studied
Acknowledgements
We would like to thank the anonymous reviewers for their valuable comments. We also wish to thank Mikael Degermark for introducing us to the problem.
Jesus Arango is a Software Engineer at Cisco Systems, where he is part of the IP engineering team that designs and develops multicast routing protocols for the IOS operating system. He holds a Ph.D. degree from the University of Arizona and a M.S. degree from Michigan State University. He has research interests in networks and operating systems. His most recent research work focuses on wireless networks, with emphasis in packet scheduling, broadcasting and header compression.
References (22)
- et al.
Contention-based forwarding for mobile ad hoc networks
Ad Hoc Networks
(2003) - J. Arango, M. Degermark, A. Efrat, S. Pink, An efficient flooding algorithm for ad hoc networks, in: Proceedings of the...
- J. Arango, A. Efrat, S. Ramasubramanian, M. Krunz, S. Pink, Retransmission and backoff strategies for broadcasting in...
- J. Arango, A. Efrat, S. Ramasubramanian, M. Krunz, Onroad vehicular broadcasting, computer communications and networks,...
- et al.
The broadcast storm problem in a mobile ad hoc network
- B. Williams, T. Camp, Comparison of broadcasting techniques for mobile ad hoc networks, in: Proceedings of the ACM...
- V.K. Paruchuri, A. Durresi, D.S. Dash, R. Jain, Optimal flooding protocol for routing in ad hoc networks, in: IEEE...
- J. Cartigny, D. Simplot, J. Carle, Stochastic flooding broadcast protocols in mobile wireless networks, Technical...
- et al.
Providing reliable and fault tolerant broadcast delivery in mobile ad hoc networks
Monet
(1999) - et al.
Dominating sets and neighbor elimination-based broadcasting algorithms in wireless networks
IEEE Transactions on Parallel and Distributed Systems
(2002)
A dominating-set-based routing scheme in ad hoc wireless networks
Telecommunication Systems
Cited by (6)
On channel-discontinuity-constraint routing in wireless networks
2014, Ad Hoc NetworksCitation Excerpt :The construction of the spanner using the above procedure can be performed using expected O((n/θ) logn) number of messages. Our proof follows the ideas of Arango et al. [39]. The first step is to show that for each node p, the expected number of messages transmitted is O((1/θ) logd) where d is the degree (number of neighbors) of p.
Stateless adaptive reliable broadcast protocol for heterogeneous wireless networks
2015, Proceedings - International Conference on Advanced Information Networking and Applications, AINAA new high performance approach: Merging optimal multicast sessions for supporting multisource routing
2013, Journal of SupercomputingRNBB: A reliable hybrid broadcasting algorithm for ad-hoc networks
2012, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)Research on collision avoidance mechanisms of broadcasting MAC protocol
2011, Proceedings - 3rd International Conference on Multimedia Information Networking and Security, MINES 2011On channel-discontinuity-constraint routing in wireless networks
2010, Proceedings - IEEE INFOCOM
Jesus Arango is a Software Engineer at Cisco Systems, where he is part of the IP engineering team that designs and develops multicast routing protocols for the IOS operating system. He holds a Ph.D. degree from the University of Arizona and a M.S. degree from Michigan State University. He has research interests in networks and operating systems. His most recent research work focuses on wireless networks, with emphasis in packet scheduling, broadcasting and header compression.
Alon Efrat is an associate professor in the Department of Computer Science at the University of Arizona. He has earned his PhD from Tel-Aviv University under the supervision of Prof. Micha Sharir. He was also a post-doctorate research assistant at Stanford University, and at IBM Almaden Research Center. His research areas include geometric algorithms and their applications to sensor networks, robotics and computer vision. He is the author or co-author of nearly 95 publications, almost all in peer-reviewed, prestigious venues. He won the NSF CAREER award in 2004. He has served on many NSF panels and technical program committees in different areas, on the editorial board of the International Journal of Computational Geometry and its Application (IJCGA), was a guest editor of this journal, and served on the TPC of many prestigious conferences including ACM GIS, FOCS, SoCG, INFOCM and others.
Srinivasan Ramasubramanian received the B.E. (Hons.) degree in Electrical and Electronics Engineering from Birla Institute of Technology and Science (BITS), Pilani, India, in 1997, and the Ph.D. degree in Computer Engineering from Iowa State University, Ames, in 2002. He is currently an Associate Professor in the Department of Electrical and Computer Engineering at the University of Arizona, where he held the position of Assistant Professor from August 2002 to July 2008. He is a co-developer of the Hierarchical Modeling and Analysis Package (HIMAP), a reliability modeling and analysis tool, which is currently being used at Boeing, Honeywell, and several other companies and universities. His research interests include architectures and algorithms for optical and wireless networks, multipath routing, fault tolerance, system modeling, and performance analysis. He has served as the TPC Co-Chair of BROADNETS 2005 and ICCCN 2008 (Optical Networking Symposium) conferences and is an editor of the Springer Wireless Networks Journal.
Stephen Pink is a Professor of Computing at Lancaster University in the UK. Prior to Lancaster, he was a Professor at Lulea University in Sweden and an Associate Professor at the University of Arizona. Pink’s interests are in protocol design and implementation, IP router design, as well as high speed and wireless networking. He has published in ACM, IEEE, IFIP and other conferences and journals over the last 20 years.
Marwan M. Krunz received the Ph.D. degree in electrical engineering from Michigan State University in July 1995. He joined the University of Arizona in January 1997, after a brief postdoctoral stint at the University of Maryland, College Park. He is currently a professor of electrical and computer engineering and the co-director of Connection One, a joint NSF/state/industry IUCRC cooperative research center that focuses on RF and wireless communication systems and networks. Dr. Krunz chaired the computer engineering group of the ECE Department (2006-2008). He previously held various visiting research positions at INRIA, HP Labs, US West Advanced Technologies, and Paris VI University. Dr. Krunz’s research is in communications technology and networking, with particular emphasis on resource allocation, adaptive control, and distributed protocol design. Recently, he has been involved in projects related to cognitive radios ; distributed resource management in wireless networks; protocol design for ad hoc networks; MIMO and smart-antenna-based systems; UWB-based wireless personal area networks; energy management and clustering in sensor networks; video streaming; and fault detection in all-optical networks. He has published more than 140 journal articles and refereed conference papers in these areas. He received the NSF CAREER Award (1998-2002). He currently serves on the editorial board for the IEEE/ACM Transactions on Networking, the IEEE Transactions on Mobile Computing, and the Computer Communications Journal. He served as a Technical Program Chair for the IEEE INFOCOM 2004, the IEEE SECON 2005, and the IEEE WoWMoM 2006 Conferences.