Opportunistic topology control for ad hoc wireless network survivability enhancement based on LIMOS model

doi:10.1016/j.jnca.2019.02.010

Journal of Network and Computer Applications

Volume 134, 15 May 2019, Pages 1-12

https://doi.org/10.1016/j.jnca.2019.02.010 Get rights and content

Abstract

Traditional topology controls existed on wireless networks are mostly based on a deterministic link model which only takes the decisive links into consideration. All these algorithms do not utilize the benefit of wide existence of lossy links. In a more realistic environment, lossy links are tremendous. Failure to utilize these links leads to low resource usage and relatively low network survivability. In this paper, we propose a lossy probabilistic network model with Obstacle Shadow awareness ability based on the quasi-UDG Link Model (LIMOS model). On the basis of this model, we design a Lossy Link Aware Opportunistic Topology Control (LOLA-OTC) algorithm which aims to utilize high quality lossy links in order to enhance the network performance. The computational complexity of the LOLA-OTC is O(Δn² + Δn) and the total message complexity of LOLA-OTC is O(n). We also present performance evaluation results on the energy consumption, packet loss ratio, and network throughput according to different node density settings through extensive simulations under ns2 network simulator. The numerical results demonstrate that the LOLA-OTC can obtain better network performance compared to other non-opportunistic topology control algorithms in most cases.

Introduction

A mobile ad hoc network (MANET) has no fixed networking infrastructure. It consists of multiple, possibly mobile, nodes that maintain network connectivity through opportunistic wireless communications. In order to improve the survivability of such network, lots of aspects should be enhanced, such as the energy saving, routing efficiency, network lifetime, etc. There are many ways to tackle these problems. One of the most widely used ideas is through topology control.

Topology control in ad hoc wireless networks aims to improve some network metrics through dynamic selection of the links among nodes. The unnecessary links can be deleted logically or physically by power control of the nodes.

In real situations, the links among the mobile nodes are usually considered to be random. Many factors may affect the quality of wireless links, such as obstacles, weather, node speed, etc. So wireless links can be unreliable in practice due to the presence of physical barriers between nodes or harsh environmental conditions (Zhao, 2017). Depending on the way of dealing with the factors, different kinds of link models have been proposed in recent years.

The model integrating all these factors is the physical model which is also called the Signal-to-Interference-plus-Noise Ratio (SINR) model (Swamy et al., 2017). Under this model, a link is thought to exist if and only if the SINR received by the receiver exceeding a threshold so that the transmitted signal can be decoded with an acceptable bit error rate (BER). Though the physical model seems to be the one closest to reality, it is not suitable for theoretical research because of its complexity.

In order to narrow the gap between theory and practice, the so-called protocol models (Ng et al., 2013; Wang et al., 2017; Kuhn et al., 2007) have been proposed in recent years. These protocol models try to simplify some reality factors to reduce the complexity. Hence, all of them are born with some limitations. In order to utilize topology control to enhance ad hoc wireless network survivability, how to decide which links should be added or deleted is the key problem that must be studied carefully. Unfortunately, due to the limitations of the underling network model, most existed topology control algorithms cannot achieve accurate results. Tradeoff between complexity and accuracy of the underling link models should be carefully considered. In recently years, more and more scholars realize that the so-called transitional region (Zuniga and Krishnamachari, 2004a, 2004b; Zhao and Govindan, 2003) phenomenon does widely exist in the ad hoc wireless networks, especially when the power of the nodes are small. In transitional region, wireless links are thought to be intermittently connected. Such links are called lossy links.

Lossy links are used to be considered as harmful when compared to the deterministic ones, because of their unreliable attributes. People believe they should be refrained from using at all times (Gao et al., 2009; Li and Guo, 2013). However, some works began to realize that the lossy links, if utilized carefully, could improve the performance of the network. Though these existed works have already investigated some effects of the lossy links, they do not analysis the underling network model and thus could be biased when compared to the real situations. So, in order to obtain more accurate research results and investigate deeper on the effect of lossy links and study the benefits of opportunistic topology control on them, the joint research of network model and topology control should be developed.

This paper first proposes a lossy probabilistic network model with Obstacle Shadow awareness ability based on the quasi-UDG Link Model, named LIMOS, and an opportunistic topology control algorithm, named LOLA − OTC, which can be used on the proposed LIMOS model. Through extensive simulations, the proposed LOLA − OTC algorithm indicates it can obtain higher network throughput and lower packet loss ratio when used with LIMOS model, compared with deterministic topology control and opportunistic topology control using the stochastic quasi-UDG link model.

The rest of the paper is organized as follows. Section 2 presents the related work of some protocol network models and topology control algorithms. Section 3 states our LIMOS network model and describes its mathematic attributes. Section 4 presents the LOLA-OTC algorithm based on LIMOS model and gives its theoretical analysis. Section 5 describes simulation results of the LOLA-OTC algorithm. Section 6 summarizes this paper.

Section snippets

Protocol model

One of the widely used protocol models is the Unit Disk Graph (UDG) model (Ng et al., 2013), as depicted in Fig. 1. In a circular disk area originate from node O with radius x, every node in this area is considered to be connected with node O. The node A within this area is thought to has a link with node O, this link is deterministic whatever the situations of the networks have. So, the link existence probability always equals to one, and the strength of link AO is always equals to other links

Formulations and models

Let us assume that we are dealing with a two-dimensional bounded region $O$ , where $O \in R^{2}$ . A finite number of spatial nodes located according to the Poisson probability distribution. $N$ is the node set where $N = {N_{1}, N_{2}, N_{k} \dots, N_{n}}$ . Each node can be seen as an independent random two-dimensional vector having common probability density function. The density function f is assumed fixed but arbitrary unless stated otherwise, subject only to the conditions that f is measurable and satisfies $\int_{R^{2}} f (x) d x = 1$ , and

Definitions and thereom

We denote $G (V, E, M)$ as a graph representing the network under a model $M$ , where $V$ and $E$ represent the vertexs and the edges in the network under model $M$ , respectively. So if an edge $E_{i j} \in E$ , the node v_i and v_j are said to be connected under the model $M$ . We denote $G (V, E {, M}_{U D G})$ as the graph representing the network under the UDG model with all edges $E_{i j} \in E$ as long as ∥v_i − v_j∥ ≤ 1. In the same way, $G (V, E {, M}_{L I M O S})$ can represents the network under the LIMOS network model with all edges $E_{i j} \in E$ as long as ∥v_i

Simulation environment

In the following sections, we compare different network metrics under different network situations using the ns2 simulator. The simulation parameters are illustrated in Table 1. Fifty nodes are placed on different size of area according to Poisson distribution as depicted in Fig. 14. The AOMDV is used as the routing protocol. The simulation time is 200 s and the propagation model is the FreeSpace model (Friis, 1946). We set CSthreshold to 6.35631e-10W and Rxthreshold to 3.07645e-9W. That means

Conclusions

In the ad hoc wireless network environment, there are a lot of devices connected with each other through deterministic links or lossy links. The lossy links tend to be tremendous according to a lot of research work. Most existing research works are based on the ideal UDG network model. Although some works use the more realistic one, such as the stochastic quasi-UDG model, they fail to take the obstacle effect into consideration and thus leading to biased research results.

In this paper, we

Acknowledgment

This work was supported by the Educational Commission of Hubei Province of China [Grant Number Q20152202] and the National Scholarship Council (CSC) of China [Grant Number 201608420175].

Tong Wang received the M.S. degree in Computer Software and the Ph.D. degree in Computer Software from Nankai University, China and Wuhan University, China in 2005 and 2012, respectively. He is an associate professor in the Department of IOT, School of information engineering, Hubei University of Economics, Wuhan, China. He was a visiting scholar with the Department of Computer Science, College of Engineering and Applied Sciences, Western Michigan University, MI, USA. His current research

References (25)

X. Ao et al.
Distributed cooperative topology control for wanets with opportunistic interference cancelation
IEEE Trans. Veh. Technol.
(2014)
M. Benter et al.
Reactive planar spanner construction in wireless ad hoc and sensor networks
S. Chiu et al.
Stochastic Geometry and its Applications. Wiley Series in Probability and Statistics
(2013)
H.T. Friis
A note on a simple transmission formula
Proc. IRE
(1946)
Q. Gao et al.
Cross-layer rate control in wireless networks with lossy links: leaky-pipe flow, effective network utility maximization and hop-by-hop algorithms
IEEE Trans. Wirel. Commun.
(2009)
B. Guo et al.
Energy-efficient topology control with selective diversity in cooperative wireless ad hoc networks: a game-theoretic approach
IEEE Trans. Wirel. Commun.
(2014)
J. Guo et al.
Distributed fault-tolerant topology control in cooperative wireless ad hoc networks
IEEE Trans. Parallel Distrib. Syst.
(2015)
M. Haenggi
On distances in uniformly random networks
IEEE Trans. Inf. Theory
(2005)
F. Kuhn et al.
Ad-hoc networks beyond unit disk graphs
Wireless Netw.
(2007)
P. Li et al.
On the multicast capacity in energy-constrained lossy wireless networks by exploiting intrabatch and interbatch network coding
IEEE Trans. Parallel Distrib. Syst.
(2013)

X. Li et al.

Hypocomb: bounded-degree localized geometric planar graphs for wireless ad hoc networks

IEEE Trans. Parallel Distrib. Syst.

(2013)

Y. Liu et al.

A generalized probabilistic topology control for wireless sensor networks

IEEE J. Sel. Area. Commun.

(2012)

Cited by (3)

Survivability evaluation for networks carrying complex traffic flows
2023, Digital Communications and Networks
The capability of a system to fulfill its mission promptly in the presence of attacks, failures, or accidents is one of the qualitative definitions of survivability. In this paper, we propose a model for survivability quantification, which is acceptable for networks carrying complex traffic flows. Complex network traffic is considered as general multi-rate, heterogeneous traffic, where the individual bandwidth demands may aggregate in complex, nonlinear ways. Blocking probability is the chosen measure for survivability analysis. We study an arbitrary topology and some other known topologies for the network. Independent and dependent failure scenarios as well as deterministic and random traffic models are investigated. Finally, we provide survivability evaluation results for different network configurations. The results show that by using about 50% of the link capacity in networks with a relatively high number of links, the blocking probability remains near zero in the case of a limited number of failures.
Joint link-level and network-level reconfiguration for urban mmWave wireless backhaul networks
2020, Computer Communications
Citation Excerpt :
Specifically, [34,35] proposed analytical methods and path restoration schemes to prevent single-link failures in wavelength-division- multiplexing optical networks. While in wireless multi-hop networks, most related works focus on the protection method, because it can achieve fast recovery [36–39]. In [36], the author developed an effective heuristic algorithm to find disjoint backup paths in survivable networks, which shows the good performance on both optimality and running time.
Due to the rapid densification of small cells in 5G and beyond cellular networks, deploying wired high-bandwidth connections to every small cell base station (BS) is difficult, particularly in older metropolitan areas where infrastructure for fiber deployment is lacking. For this reason, mmWave wireless backhaul has been proposed as a cost-effective and flexible alternative that has the potential to support the high data rates needed to accommodate backhaul traffic demands. To address the robustness of such networks, we investigate a novel relay-assisted backhaul architecture, where a number of small-cell BSs and relays are deployed, e.g. on the lampposts of urban streets. In this scenario, the interconnected logical links constitute a mesh network, which offers opportunities for both link-level and network-level reconfiguration to overcome blockages and/or node failures. We present two joint link-network level reconfiguration schemes for recovery after exceptional events. One prioritizes relay path (link-level) reconfiguration and uses alternate network-level paths only if necessary. The other splits traffic on both reconfigured logical links and backup network paths to improve throughput. Through simulation, the reconfiguration schemes are shown to not only provide near-optimal backhaul survivability but to also maintain high network throughput across a range of scenarios for urban mmWave backhaul networks. The schemes are also validated to significantly outperform existing purely link-level and purely network-level reconfiguration schemes.
Topology Control Algorithms for Wireless Sensor Networks: A Review
2020, Wireless Personal Communications

Leszek T. Lilien is an associate Professor of Computer Science at Western Michigan University (WMU) and Adjunct Associate Professor of Computer Science at Purdue University (2015, 2013).Ph.D. in Computer Science, University of Pittsburgh. M.S. in Electronics Computer Engineering, Wroclaw University of Technology, Wroclaw, Poland. Research at Western Michigan University (WMU) as a faculty; Purdue University as a post doctoral researcher, a sabbatical visitor and an adjunct faculty; and University of Illinois at Chicago (UIC)-as a faculty. Teaching at WMU and UIC. Tutorial instructor for IEEE Computer Society. Diversified RD experience at ATT Lucent Bell Labs. Consultant for academic and business projects. Entrepreneurial experience in the United States and Poland.Current research focused in two Areas: (1) opportunistic capability utilization networks or oppnets (a specialized kind of ad hoc networks); and (2) trust, privacy and security in open computing systems.

View full text

Opportunistic topology control for ad hoc wireless network survivability enhancement based on LIMOS model

Abstract

Introduction

Section snippets

Protocol model

Formulations and models

Definitions and thereom

Simulation environment

Conclusions

Acknowledgment

Distributed cooperative topology control for wanets with opportunistic interference cancelation

IEEE Trans. Veh. Technol.

Reactive planar spanner construction in wireless ad hoc and sensor networks

Stochastic Geometry and its Applications. Wiley Series in Probability and Statistics

A note on a simple transmission formula

Proc. IRE

Cross-layer rate control in wireless networks with lossy links: leaky-pipe flow, effective network utility maximization and hop-by-hop algorithms

IEEE Trans. Wirel. Commun.

Energy-efficient topology control with selective diversity in cooperative wireless ad hoc networks: a game-theoretic approach

IEEE Trans. Wirel. Commun.

Distributed fault-tolerant topology control in cooperative wireless ad hoc networks

IEEE Trans. Parallel Distrib. Syst.

On distances in uniformly random networks

IEEE Trans. Inf. Theory

Ad-hoc networks beyond unit disk graphs

Wireless Netw.

On the multicast capacity in energy-constrained lossy wireless networks by exploiting intrabatch and interbatch network coding

IEEE Trans. Parallel Distrib. Syst.

Hypocomb: bounded-degree localized geometric planar graphs for wireless ad hoc networks

IEEE Trans. Parallel Distrib. Syst.

A generalized probabilistic topology control for wireless sensor networks

IEEE J. Sel. Area. Commun.