Scheduled-links multicast routing protocol in MANETs

https://doi.org/10.1016/j.jnca.2015.12.013Get rights and content

Abstract

Dynamically changing of network topology due to node movement is considered a main challenge in providing reliable and low-cost paths in one to many ends of communication in mobile ad hoc networks. To address this challenge, this paper proposes a scheduled-links multicast routing protocol (SLMRP) based on mobility prediction. SLMRP constructs multiple scheduled paths between multicast sources and receivers. SLMRP scheduled paths are subject to reliability and quality of service requirements in load-balance strategy. Multiple loop-free and node-disjoint paths are constructed for each source–receiver pair during route discovery process. To control the activation and deactivation of these paths, we introduce multicast routing activation timer (MRAT) and path timeout timer (PTT). MRAT and PTT are computed according to the route expiration time for the set of the discovering paths. Performance comparison results demonstrate that the proposed protocol outperforms PUMA, MAODV, and PMRP.

Introduction

Mobile ad hoc networks (MANETs) are autonomous systems that comprise a collection of mobile nodes that use wireless transmission for communication. They are self-organized, self-configured, and self-controlled infrastructure-less networks (Sarkar et al., 2008). Dynamic changing of network topology due to node movement is considered the most prominent feature of MANETs (Younis et al., 2006, Gup et al., 2007). In addition, the topology of an ad hoc network can vary because of the characteristics of the radio channels. The limited radio propagation range of wireless devices imposes that the communication in MANETs is often multihop, which causes the node to work as a host or router. Moreover, bandwidth and battery power are limited. These constraints together with the dynamic network topology, make routing and multicasting in ad hoc networks extremely challenging (Bae et al., 2000). As a result of the infrastructure-less nature of mobile networks, as well as their flexibility and low cost, such networks have had widespread applications in recent years, such as disaster recovery, rescue missions, military operations in a battle-field, conferences, and outdoor entertainment activities. One-to-many or many-to-many communications is one of the most important natures of these applications. To use unicast transmission in such applications imposes multiple signaling, which has high cost in MANETs. Instead, multicast approach is the highly preferred choice, in which, one signaling to many receptions can efficiently save node resources and network bandwidth. Therefore, multicast routing can be applied efficiently to construct group communications to deliver messages from a sender to multiple receivers. The difference between the natures of traditional wired and wireless networks prevents the application of the same multicast routing protocols. For this purpose, many multicast routing protocols have been designed to meet MANET characteristics.

Generally, multicast routing protocols can be classified based on how the routing structure is constructed and maintained; these classifications are tree-based and mesh-based routing protocols (Luo et al., 2009). Tree-based routing protocols construct either a shared-tree that connects all multicast members or a source-based tree that is established and maintained for each source node of a multicast group. In tree-based protocols, only a single path will be available to serve data packets that are forwarded from source to receiver. However, the mesh-based approach provides robustness in the case of node mobility, but it requires increasing the flooding through a mesh. This condition results in inefficient bandwidth usage, high overhead, and lack of management of traffic load. The tree-based approach is considered efficient in terms of traffic distribution and has a low overhead. The disadvantage of the tree-based approach is low reliability under high speed, in which link failures trigger a reconfiguration of the entire tree as mobility increases (Biradar and Manvi, 2012). To apply an efficient approach in terms of traffic distribution and robustness, we propose a scheduled-links mechanism for multicast routing (SLMR) in MANETs. SLMR is an improvement approach to source tree-based routing. To achieve robustness and reliability in the presence of node mobility, SLMR provides multiple node-disjoint paths between a source–receiver pair. Instead of using all these paths simultaneously, where they may interfere with each other, a scheduling scheme has been introduced to support efficient usage of network resources. As a result of applying scheduling for discovered paths, a load-balance metric has been derived from each path, which enables management of traffic load across scheduling paths.

The main contribution of this paper is to propose a new multicast routing protocol based on mobility prediction in MANETs. In this new approach, one control signaling will be used to construct and schedule multiple paths to the receiver. The routes to serve data packet forwarding are discovered and scheduled based on the cooperation process between sources and receivers. The set of discovering paths will be scheduled to support load balance and traffic distribution between these paths, as well as reduce congestion and save network resources. Load balance and traffic distribution is achieved in SLMRP through controlling path utilization time for each source–receiver pair. Path utilization time is controlled by computing multicast routing activation timers (MRATs) and path timeout timers (PTTs) according to the set of route expiration time for their paths. To increase reliability and robustness, which relies on the set of discovering paths, SLMRP route discovery mechanism ensures that these paths are loop-free and node-disjoint.

The rest of this paper is organized as follows:

Section 2 highlights some multicast routing protocols, where numerous approaches have been proposed to address and overcome obstacles during the implementation of multicast routing protocols in MANETs. Section 3 presents system models in the SLMR algorithm. Section 4 shows the operation of SLMRP, the route establishment, and maintenance. Section 5 analyzes the SLMR algorithm and proves that SLMRP discovered paths are loop-free and node-disjoint. Section 6 describes the performance comparison results between SLMRP, MAODV (Royer and Perkins, 1999), PUMA (Vaishampayan and Garcia-Luna-Aceves, 2004, and PMRP (Wang et al., 2007). Our comparison evaluates the experiments to test the performance of our proposed scheme against that of three different approaches in multicast routing, namely, mesh-based, tree-based, and mobility prediction. The comparison study uses different metrics to examine the effect of node mobility and traffic load on protocol performance. Finally, Section 7 presents the conclusion.

Section snippets

Related work

Up to date, many multicast routing protocols have been proposed and implemented for MANETs. The multicast ad hoc on demand distance vector protocol (MAODV Royer and Perkins, 1999) is considered a classical example of a shared tree-based routing protocol. MAODV is an extension of ad hoc on demand distance vector (AODV, Perkins and Royer, 1999) unicast protocol. MAODV utilizes a similar discovery process of AODV to acquire routes toward the group on demand. The protocol provides and maintains a

Mobility prediction and packet formats

Each node unitizes the location information of the corresponding GPS (Kaplan, 1996) to predict mobility parameters (i.e., location coordinates, speed, and moving direction). Two important assumptions are considered in this model. The first is a free-space propagation model (Rappaport, 1995), in which the signal strength depends solely on the distance to the transmitter. Also, all nodes are synchronized with the GPS clock in the second assumption. Assume that nodes A and B are within the same

Overview

In this section, we propose a SLMRP based on mobility prediction. SLMRP is an improvement of source tree-based routing, which is considered efficient in terms of traffic distribution. The source tree-based multicast routing is established and maintained for each source node of a multicast group. Multicast packet is forwarded along the most efficient path from the source node to each multicast group member. By contrast, mesh-based routing uses periodic signaling to establish multiple routes from

Analysis

We show that no loops can be found during the node forwarding process. At the same time, discovered routes in SLMRP are node-disjoint. Let G(V,E) be an undirected weighted graph, where V is the mobile nodes and E is the wireless links between them. The link cost equals the link LET between the connected nodes pair.

Definition 5.1

Paths Pdis1={s,a1,a2,d} and Pdis2={s,b1,b2,,d}, which connect the source s to destination d are node-disjoint paths if they do not have any common node, except the source and

Performance evaluation

We compare the performance of SLMRP against the performance of PUMA (Vaishampayan and Garcia-Luna-Aceves, 2004), MAODV (Royer and Perkins, 1999), and PMRP (Wang et al., 2007), which are state-of-the-art multicast routing protocols for MANETs. PUMA and MAODV are both receiver-based protocols. However, PUMA is a mesh-based protocol and provides multiple routes from multicast sources to receivers. By contrast, MAODV is a tree-based, multicast extension to AODV (Perkins and Royer, 1999) unicast

Conclusion

In this paper, we proposed SLMRP based on mobility prediction for MANETs. We used location information to select and schedule multipath set with auto activation and deactivation relative to paths׳ RETs to support a reliable and low-cost forwarding multicast data packets to receivers. The SLMRP discovery process ensures that the set of discovered paths is loop-free and nodes-disjoint paths, which increase reliability and protocol performance under different operation scenarios. Load balance

Acknowledgments

This work was supported by the National Key Technology Research and Development Program of the Ministry of Science and Technology of China under Grant no. 2012BAH93F01, the Innovation Research Fund of Huazhong University of Science and Technology under Grant no. 2014TS095 and the National Science Foundation of China under Grant no. 60803005.

References (22)

  • Rajashekhar C. Biradar et al.

    Review of multicast routing mechanisms in mobile ad hoc networks

    J Netw Comput Appl

    (2012)
  • S. Guo et al.

    Energy-aware multicasting in wireless ad-hoc networksa survey and discussion

    Comput Commun

    (2007)
  • Mohammad-Amin Kharraz et al.

    On-demand multicast routing protocol with efficient route discovery

    J Netw Comput Appl

    (2012)
  • Asif HM, Sheltami TR, Shekshouki EE. Power consumption optimization and delay minimization in MANET. In: Proceedings of...
  • S. Bae et al.

    The design, implementation, and performance evaluation of the on-demand multicast routing protocol in multihop wireless networks

    IEEE Netw

    (2000)
  • Das SK, Manoj BSBS, Murthy CSR. A dynamic core based multicast routing protocol for ad hoc wireless networks. In:...
  • Fall K, Varadhan K, editors. The NS manual. Available:...
  • J.J. Garcia-Luna-Aceves et al.

    The core-assisted mesh protocol

    IEEE J Sel Areas Commun

    (1999)
  • J.J. Garcia-Luna-Aceves et al.

    PRIMEan interest-driven approach to integrated unicast and multicast routing in MANET

    IEEE/ACM Trans Netw

    (2011)
  • Ghasemi M, Bag-Mohammadi M. Level-based on-demand multicast routing protocol for mobile ad hoc networks. In: 6th joint...
  • Lee S-J, Gerla M, Chiang C-C. On-demand multicast routing protocol. In: Proceedings of the IEEE WCNC, vol. 3; 1999. p....
  • Cited by (27)

    • Neighbor Knowledge-based Rebroadcast algorithm for minimizing the routing overhead in Mobile Ad-hoc Networks

      2019, Ad Hoc Networks
      Citation Excerpt :

      The routing overhead connected with the distribution of data packets such as RREQ packets could be relatively large, particularly once the network topology recurrently modifies [29]. Conventional protocols [2] constructed a huge quantity of network traffic by randomly transmitting the data packets to the whole network with RREQ packets throughout route discovery [34]. The problem of the routing overhead connected with route discovery process and maintenance process was reduced.

    • Novel strategies for path stability estimation under topology change using Hello messaging in MANETs

      2019, Ad Hoc Networks
      Citation Excerpt :

      Transmission rate of Hello messages, AMP, probability of successful reception of a Hello message, and CH selection criteria are important parameters of interest for the presented adaptive routing model in the proposed PDA strategy, which can effect dynamic link connectivity among neighboring nodes inside the cluster. In addition, these parameters of interest are helpful in investigating the impact of different performance metrics, in terms of link excess life and uniform distribution of path durations, on the stability of dynamic link among neighboring nodes inside cluster [37]. Furthermore, these performance metrics are helpful in analyzing the path duration statistics in terms of throughput and overhead.

    • Multi-constraints link stable multicast routing protocol in MANETs

      2017, Ad Hoc Networks
      Citation Excerpt :

      Moralism [13] is an extended version of LSMRP. Various QoS aware routing metrics have been proposed to determine a stable link [14–17]. The ultimate goal of all stability based protocols is to discover the long-lived routes; the difference lies in the Algorithms to estimate stability.

    • Ad Hoc Multicast Routing Protocol Utilizing Increasing ID Numbers in MANETs

      2021, EAI/Springer Innovations in Communication and Computing
    View all citing articles on Scopus
    View full text