11-05-2012, 10:36 AM
Simulations of Large-scale WiFi-based Wireless Networks: Interdisciplinary Challenges and Applications
0807.1475.pdf (Size: 218.22 KB / Downloads: 33)
Introduction
Modern world has become increasingly mobile. As a result, traditional ways of
connecting users to the Internet (and to each other) via physical cables have
proved inadequate. Wireless communications [1], on the other hand, poses no
restrictions on the user’s mobility and allows a great deal of flexibility, both on
the part of users and service providers. Wireless connectivity for voice via mobile
telephony made it possible for people to connect to each other regardless
of location. This has had a profound influence on the business of telecommunications,
as well as the society as a whole [2]. New wireless technologies
targeted at computer networks promise to do the same for Internet access,
connecting wirelessly not only laptops and portable devices but also millions
of cars, sensors, consumer devices, etc to each other and to the global Internet.
The most successful, and fastest growing, example of such wireless technologies
is WiFi (Wireless Fidelity)[3]. Like cellular technology, WiFi uses a number of
base stations to connect user devices to an existing fixed network (these base
stations are called access points). However, unlike cellular systems which are
centralised, WiFi systems operate in a highly distributed fashion. Each WiFi
device is responsible for managing its own connectivity, mobility and access to
the radio spectrum. Furthermore, unlike cellular systems, nearby WiFi devices
can directly connect to each other and form self-organising wireless adhoc
networks [4,5]. Such networks are highly dynamic and flexible. They can be
created (and torn down) on the fly in order to route data packets between
participating devices, or to the closest Internet gateway. Adhoc technology can
also be used to connect together a collection of WiFi accesspoints which then
form a so-called mesh network [6]. This can help to greatly extend the range
of WiFi coverage without the need for connecting every single accesspoint to
the fixed network.
Modelling Ingredients
There has been significant previous research in modelling [12] and simulations
of wired communication networks [10,11]. However, modelling of wireless networks
is very distinct from modelling of wired networks in that the physical
medium properties, i.e. radio propagation and interference, cannot be separated
from the higher layer network protocols, because strong interactions
impact performance and drive engineering design decisions. Furthermore the
ability of users to (rapidly) change their physical location while maintaining
connectivity greatly increases the dynamism of these networks, in comparison
to fixed networks. In this section we shall focus on describing these distinctive
ingredients for the modelling of WiFi-based wireless networks. Other components
in high-fidelity simulations of these networks, which are outside the scope
of the current paper include modelling of various packet routing mechanisms
[1,14] and transport protocols [13] in WiFi environments.
Graph representation of interactions in WiFi Networks
From the above models of radio propagation, an abstract communication
graph for a collection of WiFi devices can be constructed. This is achieved
by creating an edge between node i and all other nodes in the plane that are
within the transmission range of i, and repeating this procedure for all nodes
in the network. In general wireless devices may use different transmit powers
such that the existence of a wireless link from i to j does not imply that a link
from j to i also exists. Consequently the resulting communication graph is
directed. Assuming, however, that all devices use the same transmit power P,
and a corresponding transmission range rt, the topology of the resulting network
can be described as a two dimensional random geometric graph (RGG)
[18]. Similarly, one constructs an interference graph for the network by creating
an edge between any two nodes which are within a radius ri of each other.
Fig 1. shows, as an examples, the communication and interference graphs created
by a collection of WiFi devices distributed randomly in a 1000 × 1000
m2 rectangular area.