TOWARDS MIDDLEWARE SUPPORT

FOR PERVASIVE COMPUTING

Dionisis X. Adamopoulos

Department of Technology Education and Digital Systems, University of Piraeus, Greece

Keywords: Mobile ad hoc networks, location-aware routing, location-aware services, Jini technology.

Abstract: Mobile ad hoc networks (manets) are dynamically reconfigurable multi-hop wireless networks with no fixed

infrastructure, consisting of radio-equipped mobile hosts. Each host acts as a router and moves in an

arbitrary manner. Under such a network environment routing becomes a challenging task that can be

significantly supported and facilitated by the exploitation of location information, as this paper argues. More

specifically, after a brief introduction to routing in manets, a location discovery algorithm is proposed.

Then, the paper focuses on location-aware routing and after presenting briefly the most important related

protocols attempts to compare them based on a number of qualitative properties. Finally, the emergence of

location aware services is discussed, a service discovery scenario based on Jini technology is proposed and

important related deployment challenges are highlighted.

1 INTRODUCTION

Mobile communication and wireless networking

attract an increasing interest due to recent

improvements in microprocessor and radio

technologies, and are becoming more and more

popular as they eliminate the disadvantage of having

the user restricted in a particular location. Wireless

systems constitute a mature solution that can be

deployed easily, decreasing the overall cost and

extending the working environment to a global one,

utilising different wireless access technologies,

interworking with each other transparently.

In most mobile communication environments

there are base stations, which directly keep track of

the mobile hosts. Although two mobile hosts are in a

transmitting range of each other, they must

communicate with each other by some base station.

However, under some circumstances (e.g., due to

low cost effect, poor performance or low usage), the

mobility support offered by base stations is not

available. In this environment the mobile hosts must

form an ad hoc network.

Broadly defined a mobile ad hoc network

(manet) is temporarily formed by a group of

autonomous wireless nodes / hosts that communicate

with each other over wireless channels and

cooperatively form a network that operate without

the support of any fixed network infrastructure and

centralized administration (Mauve, 2001).

Routing in manets affects directly the efficiency

of the network and is more challenging than in

traditional networks because of the dynamic

character of the network topology. In the last few

years, a large number of routing protocols have been

proposed for manets (Royer, 2004). Two different

routing approaches can be distinguished: topology-

based and position-based (or location-aware)

routing. For this reason, this paper after discussing

important matters related to the enabling of location

awareness, examines the exploitation of location

information in manets, including ways that this

information can assist manet routing and methods

for providing and supporting location-aware

services.

2 ENABLING LOCATION

AWARENESS

A widely accepted theoretical model for manets is

the unit graph model, which represents the mobile

hosts of the manet spread out in some environment

by a set of points in the Euclidean plane. According

to this model two nodes A and B in the network are

neighbors (and thus joined by an edge) if the

287

X. Adamopoulos D. (2007).

TOWARDS MIDDLEWARE SUPPORT FOR PERVASIVE COMPUTING.

In Proceedings of the Ninth International Conference on Enterprise Information Systems - SAIC, pages 287-290

DOI: 10.5220/0002390802870290

 SciTePress

distance between them is at most R, where R is the

transmission radius which is equal for all nodes in

the network and specifies the maximum (Euclidean)

distance between two mobile hosts at which they are

able to directly communicate (Stojmenovic, 2004).

The use of the nodes’ position for routing

purposes in a manet represented as a unit graph,

assumes that each node is aware of its own position

(its x- and y- coordinates) in the plane. The distance

between neighboring nodes can be estimated on the

basis of incoming signal strengths or time delays in

direct communications. Relative coordinates of

neighboring nodes can be obtained by exchanging

such information between neighbors (Capkun,

2005). Alternatively, the location of nodes may be

available directly by communicating with a satellite,

using GPS (Global Positioning System), if nodes are

equipped with a small low power GPS receiver.

Independent of the exact positioning technology

used location-aware routing protocols make

forwarding decisions based on the geographical

position of a packet’s destination. Therefore, the

main prerequisite for location-aware routing is that a

sender can obtain the current position of the

destination, assuming that it already knows its own

position. In order to satisfy this need a location

discovery algorithm is proposed, which can be used

as is in simple routing scenarios or adjusted and in

an integrated manner in the framework of various

position-based packet forwarding strategies and

location-aware routing protocols (Mauve, 2001).

It is assumed that the geographical area of a

manet is divided into rectangular regions, as can be

seen in Figure 2. All of these regions have well-

defined IDs that are concatenations of the x and y

coordinates of the bottom-left and upper-right

corners. Furthermore, all nodes have a unique ID

(such as an IP address). Next, it is assumed that

there exists a static mapping f that maps a node’s ID

into a specific region (called its home region):

f(Node ID) Æ Region ID

The function f represents a many-to-one mapping

that is static and known to all nodes of the manet. It

is necessary to satisfy the following properties:

•

Every region should have the same node density in

order to evenly distribute the queries throughout

the manet.

• The entry or the departure of nodes from the

manet should be transparent to f.

• The shape or size of the geographical area covered

by the manet should not affect f.

Under these conditions the proposed location

discovery algorithm, which is executed by every

node in a manet, can be represented using pseu-

docode in the following way:

LocationDiscoveryAlgorithm

{

if (the current node wants to move) then

{

Move the Current Node (CN);

if (the CN moves out of its current

region) then

{

Use function f to find the home

region of the CN;

Broadcast a location update message to

all nodes in the home region of the CN;

Inform all nodes in the home region of

the CN about the new location of the CN

- its new region;

}

if (the CN wants to send a packet to

another node) then

{ // determine the current location of the

// recipient node

Use function f to find the home region of

the recipient node;

Send a message to this region enquiring

about the current location of the

recipient node;

Wait for the desired respond;

Send the packet to the target node

(e.g. using a location-aware routing

protocol);

}

A function f that can be used together with the

proposed location discovery algorithm and satisfies

all the necessary criteria is:

f(ID) = g(ID) mod k,

where g(ID) Æ [0, …,N] is a random number

generating function that uses the node ID as a seed

and outputs a random number. The number of home

regions that are present in the manet is denoted by k.

3 LOCATION-AWARE ROUTING

In a manet each node communicates directly with

nodes within wireless range and indirectly with all

other destinations using a dynamically determined

multi-hop path via other nodes in the network. In

this multi-hop scenario several hosts may need to

relay a packet before it reaches its final destination.

Therefore, the task of finding and maintaining routes

in manets in order to send a message from a source

node to a destination node is nontrivial because of

the constantly and dynamically changing topology

of the network triggered by node mobility.

ICEIS 2007 - International Conference on Enterprise Information Systems

288

One prospective approach to assist routing in a

manet leverages the data available from location

information. Since it is not necessary to establish or

maintain explicit routes, the nodes have neither to

store routing tables nor to transmit messages to keep

routing tables up to date, and thus location-aware

routing does scale well even if the network is highly

dynamic. This is a major advantage in a manet

where the topology may change frequently.

The location information that mobile hosts in

manets provide can exploit the geometric relation-

ship among those hosts optimising significantly the

routing process. The most important manet routing

protocols that feature location awareness are the

following (Mauve, 2001)(Tseng, 2001): Location-

Aided Routing (LAR), Distance Routing Effect

Algorithm for Mobility (DREAM), Geographic

Distance Routing (Gedir), Grid and Zone-based two-

level routing.

Table 1: Comparison of location-aware routing protocols.

Protocol

Name

Path

Strategy

Loop

Freedom

Performance

Metrics

Memory

LAR Flooding No Hop count Yes

Gedir Single path No Hop count No

Grid Single path Yes Hop count Yes

Zone-

based 2-

level

Single path Yes Hop count No

DREAM Flooding No Hop count Yes

Power

aware

Single path Yes Power No

Cost

aware

Single path Yes Cost No

Power-

cost aware

Single path Yes Power, cost No

Protocol

Name

Distributed

operation

Scalability

Guaranteed

Delivery

Robustness

LAR Localised No No No

Gedir Localised Yes No No

Grid Localised No No No

Zone-

based 2-

level

Zonal No Yes No

DREAM Localised No No No

Power

aware

Localised Yes No No

Cost

aware

Localised Yes No No

Power-

cost aware

Localised Yes No No

Location-aware routing protocols can be

compared according to the following qualitative

properties / characteristics (Macker, 2004) (Stojme-

novic, 2004): Path strategy, loop freedom,

performance metrics, memorization, distributed

operation, scalability, guaranteed message delivery

and robustness:

Table 1 compares important existing location-

aware routing protocols based on the proposed

qualitative properties / characteristics. Some of these

protocols are fully location aware (e.g., Grid) as they

exploit location information in route discovery,

packet relay and route maintenance. Some others are

only partially location aware (e.g., LAR is location

aware only in terms of route discovery). Table 1

includes two power / cost aware routing protocols

(Royer, 2004), because of their importance. While

the computational power of the devices used in the

network is rapidly increasing, the lifetime of

batteries is not expected to improve much in the

future. From this table is also obvious that

robustness is not inherently supported by current

location-aware routing protocols.

4 LOCATION-AWARE

TELECOMMUNICATIONS

SERVICES

The available location-tracking technologies and the

position-based routing protocols, combined with the

massive adoption of pervasive computing devices

enable location-aware services in manets, forming

an important and very promising class of mobile

commerce applications..

In a manet environment, location-aware services

and other more general telecommunications services

(e.g., name resolution, file system management,

mail, Web services) cannot be centralized, because

the network population and topology are not known

in advance. Nor can they be preconfigured, since

much of the configuration will not be determined

until the network is instantiated and may need

frequent updating due to its unpredictable nature.

Therefore, service discovery is an important compo-

nent for ad hoc communications and collaboration in

ubiquitous computing environments, since it enables

the participating entities to provide services to peers

and to be aware of and use the available services

from peers.

In order to enable services to be offered and

found in a manet environment the use of Jini

technology (Kumaran, 2006) is proposed. More

specifically, the desired functionality originates by

the combined use of three Jini protocols called,

discovery, join and lookup. The discovery protocol

is used when a service is in need of a lookup server

to register. The join protocol is used when a service

TOWARDS MIDDLEWARE SUPPORT FOR PERVASIVE COMPUTING

289

has located a lookup server and wishes to join it. The

lookup protocol is used when a client / user needs to

locate and invoke a service described by its interface

type (written in Java) and possibly, other attributes.

Figure 1: Service discovery in a manet.

The proposed service discovery scenario is

illustrated in Figure 1. Initially, a service has to be

added in a Jini system. The service provider, which

is the originator of the service, locates a lookup

server by multicasting a request on the manet for any

lookup server to identify itself (discovery process in

Figure 1). Then, a service object corresponding to

the service is loaded into the lookup server (join

process in Figure 1). This service object contains a

Java interface for the service including the methods

that clients will invoke to execute the service, along

with any other descriptive attributes. The service is

now ready to be looked up and used. A client locates

an appropriate server by its type (interface and

descriptive attributes). Then, the service object is

loaded into the client to be invoked.

This service discovery approach needs optimiza-

tion in order to encounter the overhead caused by the

dynamic network topology of the manet. Explicit

multicast packets for service discovery are necessary

to be sent frequently in addition to the multicast

control packets. In this way, bandwidth and battery

resources are wasted and extra traffic is caused.

5 CONCLUSIONS

The recent availability of efficient GPS receivers

and the improvement of the techniques for finding

relative coordinates based on signal strengths, to-

gether with the need for the design of power

efficient and scalable networks, provided

justification for applying location-aware routing

methods in ad hoc networks. However, the search

for new such routing methods that have excellent

delivery rates, short hop counts, small flooding

ratios and power efficiency is far from over.

On the other hand, location-aware telecommuni-

cations services promise enhanced end-user experi-

ence and new commercial opportunities. However,

further research is necessary for providing infra-

structural support (mostly at the middleware level)

in order to alleviate the need for services to handle

issues like service discovery, privacy, context

awareness, personalization and the constrained

interfaces available on mobile devices in an ad hoc

manner, and to simplify service creation.

The field of mobile ad hoc networks is rapidly

growing and changing, and while there are still

many challenges that need to be met (collective

communication, QoS support, power-aware routing,

efficient location updating congestion avoidance,

improving network capacity), it is likely that such

networks and the corresponding telecommunications

services will see widespread use within the next few

years.

REFERENCES

Capkun, S., Hamdi, M., Hubaux, J.P., 2005. GPS-free

Positioning in Mobile Ad-Hoc Networks. In Proceed-

ings of HICSS 2005.

Kumaran, S.I., 2006. Jini Technology: An Overview,

Prentice Hall PTR, 2006.

Macker, J.P., Corson, M.S., 2004. Mobile Ad Hoc Net-

working and the IETF. In Mobile Computing & Com-

munications Review, Vol. 2, No. 1, pp. 9-14.

Mauve, M., Widmer, J., Hartenstein, H., 2001. A Survey

on Position-Based Routing in Mobile Ad Hoc

Networks. In IEEE Network, pp. 30-39.

Royer, E., Toh, C.-K., 2004. A Review of Current Routing

Protocols for Ad Hoc Mobile Wireless Networks. In

IEEE Personal Communications, pp. 46-55.

Sadeh, N., 2006. M-Commerce: Technologies, Services,

and Business Models, John Wiley & Sons.

Stojmenovic, I., 2004. Position-Based Routing in Ad Hoc

Networks. In IEEE Communications Magazine, pp.

128-134.

Tseng, Y.-C., Wu, S.-L., Liao, W.-H., Chao, C.-M., 2001.

Location Awareness in Ad Hoc Wireless Mobile

Networks. In IEEE Computer, pp. 46-52.

ICEIS 2007 - International Conference on Enterprise Information Systems

290