Network Reconﬁguration for Wireless Sensor Networks using

UML/MARTE Proﬁle

Raoudha Saida

1,2

, Yessine Hadj Kacem

, M. S. BenSaleh

and Mohamed Abid

1,2

CES Laboratory, National Engineering School of Sfax, Sfax, Tunisia

Digital Research center of Sfax (CRNS), Sfax, Tunisia

College of Computer Science, King Khalid University, Abha, Saudi Arabia

National Electronics, Comm. and Photonics Center, King Abdulaziz City for Science and Technology, Riyadh, Saudi Arabia

Keywords:

WSN, Reconﬁguration, Energy Efﬁciency, MARTE, MDE.

Abstract:

The required autonomy and the growth of the complexity of wireless sensor networks (WSNs) systems give the

reconﬁguration a big importance. Thus, the integration of reconﬁguration scenarios (node level or network

level reconﬁguration) in the development cycle of WSNs system is required to offer an efﬁcient reaction to

environment variability. In this direction, existing model based design approaches of reconﬁgurable WSNs

are limited to the design of behavioral reconﬁguration techniques which bring adjustment of the system’s

parameters or update some functionalities. However, network reconﬁguration which presents an important

ability of networked systems for dealing with the network disconnectivity and power consumption, is still

under-explored. In this context, we propose a high level model based design of WSNs applications using the

UML/MARTE standard to specify network reconﬁguration semantics. We deﬁne a new package named «NW_

Reconﬁguration». A case study on water distribution network is proposed to evaluate our proposed design

approach.

1 INTRODUCTION

A WSN (Mottola and Picco, 2011) consists of a large

number of tiny, battery powered sensor nodes that inte-

grate sensing, data processing and wireless communi-

cation capabilities. WSNs have many different appli-

cation domains such as health monitoring (Al Ameen

et al., 2012), building monitoring (Benatia et al., 2014),

environmental monitoring (de Lima et al., 2010), etc.

Despite of many different abilities, sensor nodes are

prone to failure in some cases due to their limited en-

ergy resources. A WSN application needs to modify

its behavior and its architecture in order to response to

the environment requirements and manage restricted

resources. Reconﬁguration is one of the interesting

topics in WSNs. Systems can evolve by either behav-

ioral or architectural reconﬁguration. In this context,

reconﬁguration approach in WSNs applications has

two different scenarios. The ﬁrst scenario is the node

level based reconﬁguration corresponding to the be-

havioral reconﬁguration. The second scenario is the

network level based reconﬁguration corresponding to

the architectural reconﬁguration. Network level based

reconﬁguration is one of the important required pro-

cess during the deployment since WSNs applications

are often deployed in inaccessible locations that re-

quire human intervention to reestablish the network

operating. Thus, reconﬁguration requires to be consid-

ered in the development cycle of WSNs system with

different scenarios (node level and network level) at

different abstraction levels. To cope with the growth

of the design complexity of WSNs systems, designers

have resorted to several design methods. The most pop-

ular paradigm is the Model Driven Engineering (MDE)

(Schmidt, 2006) which permits the modeling of the be-

havioral and the structural of the system and the speci-

ﬁcation of the non functional properties such as time

constraints and energy. In particular, the Uniﬁed Mod-

eling Language (UML) proﬁles allow the addition of

semantics details in ﬁrst steps of the design process. It

presents an attractive solution to support the whole life

cycle co-design of complex embedded systems with a

real time constraints and performance issues. In this

context, the Modeling and Analysis of Real Time and

Embedded systems (MARTE) (Group, 2011) proﬁle

offers a rich support for the analysis and speciﬁcation

of embedded systems. Proposed reconﬁguration mod-

els based on MARTE proﬁle are limited to deﬁning

Saida, R., Kacem, Y., BenSaleh, M. and Abid, M.

Network Reconﬁguration for Wireless Sensor Networks using UML/MARTE Proﬁle.

DOI: 10.5220/0006316002030209

In Proceedings of the 12th International Conference on Evaluation of Novel Approaches to Software Engineering (ENASE 2017), pages 203-209

ISBN: 978-989-758-250-9

203

speciﬁcations of the behavioral reconﬁguration sce-

nario which consists in adjusting system’s parameters

or updating functionalities. Indeed, there is an absence

of explicit semantics in MARTE to support the network

level based reconﬁguration. Since this reconﬁguration

scenario is not supported by MARTE proﬁle, we need

new extensions in order to cover this lack and provide

more efﬁcient design. In our work, we focus only

on the architectural reconﬁguration which consists in

modifying the topology such as adding or removing

components or connections.We aim to offer a high

level model based on the UML/MARTE standard for

modeling the speciﬁcations of this important reconﬁg-

uration scenario required in a WSNs application. The

rest of paper is structured as follows. In Section 2, we

review some related works that address reconﬁgura-

bility in complex embedded systems. In Section3, we

give an overview about modeling concepts of MARTE

standard for reconﬁgurable WSNs systems. In section

4, we describe the adopted network reconﬁguration

design strategy. In Section 5, we present in details

our proposed package to specify network reconﬁgu-

ration in WSNs. As a proof of concept we consider,

in Section 6, a case study about reconﬁgurable water

distribution networks. Finally, Section 7 concludes the

paper and presents future works.

2 RELATED WORKS

In this section, we give an overview about existing re-

conﬁguration approaches in WSNs. Many works (Vi-

dal et al., 2011), (Quadri et al., 2009), (Krichen et al.,

2011), (Cherif et al., 2011), (Said et al., 2013), (Vidal

et al., 2010) have been carried out to the modeling

of embedded systems and particularly reconﬁgurable

ones. Most of the proposed approaches were interested

in dynamic and partial reconﬁguration. For example,

a SoC co-design methodology based on MARTE stan-

dard is introduced in (Quadri et al., 2009). The re-

conﬁguration behavior is deﬁned using the speciﬁc

control semantics under GASPARD (Gamatié et al.,

2008) framework. The reconﬁgurable system is de-

scribed by a mode automata which is composed of a

state graph and a mode switch component.The state

graph describes the behavior of the system via state

based technique where states are connected with tran-

sitions. A mode switch component is constituted of a

set of modes. It enhances a switch functionality which

deﬁnes the following execution mode. In this work,

the developed model is transformed from model level

speciﬁcations to an executable FPGA platform. This

work introduces the modeling of hardware reconﬁg-

uration. In (Krichen et al., 2011), authors proposed

a model based approach which deals with the recon-

ﬁguration issues for Distributed Real time Embedded

(DRE) systems. They introduced a solution for recon-

ﬁguring DRE systems using a non predeﬁned number

of conﬁgurations. A set of mode structures are used

to model the reconﬁgurable DRE. Mode structures are

connected with transitions. Each transition presents a

reconﬁguration activity. Each node structure has var-

ious instances named modes. Every mode has a con-

ﬁguration. The proposed conﬁguration approach de-

scribes a conﬁguration by a number of structured com-

ponents, relations between them and their allocation

on the execution platforms. This work has addressed

the design of software reconﬁguration in embedded

systems. In (Said et al., 2013), authors proposed a

model based design of adaptive real time embedded

systems which supplies the MARTE standard with new

semantics to model application ﬁne-grain reconﬁgura-

tion. This work adds extensions to MARTE proﬁle in

order to support ﬁne grain adaptation. The proposed

meta model describes the behavior and structure of a

reconﬁgurable software resource. The reconﬁguration

behavior consists in a switch between a set of modes

within mode transitions. A mode can be composed of

multiple Elementary Modes connected by Elementary

Mode Transitions. Each elementary mode is described

by an output quality level and a conﬁguration parame-

ter.

All the reviewed works have proposed high level

approaches to deal with the behavioral reconﬁguration.

Although most of embedded systems, in particular

WSNs systems, have already adopted reconﬁguration

techniques in their design, there still a great need for

a WSN system to integrate the architectural recon-

ﬁguration. Indeed, the architectural reconﬁguration

modeling using MARTE proﬁle is not tackled in litera-

ture. There is no details about network reconﬁguration

speciﬁcations in WSN applications. MARTE standard

offers only features to deﬁne the behavioral reconﬁg-

uraions of embedded systems. What makes our work

different is that we aim at deﬁning new package based

on MARTE proﬁle to support architectural reconﬁgu-

ration in WSN applications.

3 PRELIMINARIES

In this section, we describe possible reconﬁguration

scenarios in WSNs application. We present then the

UML/MARTE proﬁle capabilities to specify the recon-

ﬁguration for embedded systems.

ENASE 2017 - 12th International Conference on Evaluation of Novel Approaches to Software Engineering

204

3.1 Reconﬁguration Scenarios in WSNs

WSNs applications require the deployment of hun-

dreds or thousands of sensor nodes in remote locations

that are often inaccessible. In order to achieve its

operating and connectivity, a wireless network sys-

tem refers to two different reconﬁguration scenarios

(Rajasekaran et al., 2014); node level based reconﬁgu-

ration and network level based reconﬁguration.

3.1.1 Node Level based Reconﬁguration

The dynamic reconﬁguration at node level is a required

process in WSNs applications. A sensor node adapts

its behavior to environment constraints and recovers

from broken network links. In this scenario, the hard-

ware reconﬁguration and software reprogramming are

exploited. Hardware reconﬁguration focuses on updat-

ing hardware platforms of sensor nodes and reprogram-

ming reconﬁguration concentrates on software update,

application layer and operating systems. In (Eronu

et al., 2013), authors survey existing reconﬁguration

approaches in WSN applications.

3.1.2 Network Level based Reconﬁguration

Network level based reconﬁguration is one of the im-

portant required process during the deployment. It

consists in modifying the topology such as removing

or adding elements or connections. Most of WSNs

are deployed in unmanageable environments where

the human intervention is required to reconﬁgure the

network. That’s why the network reconﬁguration is re-

quired to maintain the network topology, avoid the net-

work dis-connectivity, increase the network’s longevity

and manage limited resources more efﬁciently. In

(Alam et al., 2014), authors proposed a network recon-

ﬁguration technique in order to achieve the topological

control in the network. In the present paper, we are

concerned with the present reconﬁguration scenario.

3.2 MARTE Capabilities for

Reconﬁguration Modeling

This sub section gives an overview about UML/-

MARTE proﬁle capabilities related to reconﬁguration

issues in the modeling of embedded systems. MARTE

proﬁle is promoted by the the Object Management

Group (OMG). This proﬁle supports the modeling,

speciﬁcation and analyzing of real time embedded sys-

tems. MARTE consists of three principal packages.

Foundation package represents the foundational con-

cepts for embedded systems modeling. It offers the

description of basic real time features such as time con-

straints, non functional properties (NFPs) and other

Figure 1: MARTE capabilities for modes and conﬁguration

modeling.

resources. The second package, the MARTE design

model package, presents hardware and software spec-

iﬁcations. The third package is the MARTE anal-

ysis model package which provides annotations for

generic basis of quantitative performance and schedu-

lability analysis. MARTE introduces some concepts

to design reconﬁgurable systems. It deﬁnes in the

Foundation package the Core Elements proﬁle which

contains the Causality sub package that permits the

description of the dynamic behavioral modeling of

embedded systems. This proﬁle deﬁnes the dynamic

behavioral reconﬁguration in terms of modes and tran-

sitions between these modes. A mode is characterized

by a particular conﬁguration. A transition represents

an event which applies the switch between different

modes. The dynamics between modes is described

using state machines composed of a set of modes

and modes transitions. The current MARTE propo-

sition for reconﬁgurable system design is illustrated

in Figure 1. MARTE standard presents only concepts

for the behavioral reconﬁguration of embedded sys-

tems, however, it still lacks explicit support for the

architectural reconﬁguration modeling of networked

systems. In the rest of paper, we propose a new pack-

age called «NW_Reconﬁguration» package based on

UML/MARTE for the modeling of architectural recon-

ﬁguration applied in WSNs applications.

4 PROPOSED DESIGN

METHODOLOGY

Designing an autonomous and efﬁcient WSN archi-

tecture with satisfying the application constraints is

a complex process. Dynamic network conﬁguration

is a great need for WSN design. In this context, we

adopt a promising approach that enables the network

Network Reconﬁguration for Wireless Sensor Networks using UML/MARTE Proﬁle

205

reconﬁguration with kept of the network connectivity

and reduces the network’s power consumption. This

technique divides the network into a set of sub net-

work areas (SNAs). The proposed approach offers an

easier way to debug network problems where SNAs

are totally independents so any problem in each SNA

can easily be isolated and resolved. It enables also

the application of different topologies in each SNA

which reduces the energy consumption.This method-

ology can be expressed in two steps. In the ﬁrst step,

we begin by the network formation. The network is

composed of three types of nodes: a single base station

(BS), routers and end devices. After the identiﬁcation

of the BS and different nodes in the network by their

IDs, the BS station identiﬁes, ﬁrst, the router which

must have the lowest Received Signal Strength Indi-

cation (RSSI) value and the greater battery level and

others nodes will be considered as end points after a

determined time. Second, this router identiﬁes the next

router and their associated end points in the highest

level of the same SNA based on the same algorithm.

Two topologies can be described; the tree topology

for the connection between BS and routers and the

star topology for the connection between routers and

end points. The second phase, is the dynamic network

reconﬁguration in which an automatic self reconﬁgu-

ration done if any change occurs in the physical topol-

ogy of the network. This reconﬁguration technique

depends on two major parameters: the RSSI value and

the battery level of nodes. The RSSI value speciﬁes the

strength of the received signal. Its value decreases with

the increase in distance between the receiver and the

sender. Battery level describes the remaining energy

enables the continuity of the node’s operations. Any

modiﬁcations in these two parameters could affect in

the assignment of different task (router or end device)

to that considered mote. The network also can detect

presence of any new mote in its coverage area and

assign its role based on the two mentioned parameters.

Likewise, the remove of any node from the network

needs a dynamic reconﬁguration and a replacement of

the removed node.

5 THE PROPOSED «NW_

RECONFIGURATION»

PACKAGE

In this section, we give a description of the new pro-

posed network reconﬁguration package. The structure

of the «NW_ Reconﬁguration» package is shown in

Figure 2. Within this proposed package, reconﬁgura-

tion properties and network features are declaratively

Figure 2: The NW_Reconﬁguration package overview.

speciﬁed. We deﬁne a number of stereotypes to re-

late Network reconﬁguration elements based on the

UML/MARTE proﬁle. Data types from MARTE Non

Functional Properties (NFPs) are used to represent

some attributes. The stereotype «GaExecHost» from

the GQAM sub proﬁle and the stereotype «HwCom-

municationResource» from GRM sub proﬁle are used

to specify others semantics of classes. We detail the

modeling of different features for the network recon-

ﬁguration in the next sub sections.

5.1 «Network» stereotype

It extends the semantics of the «Network» proﬁle

proposed in (Ebeid et al., 2015).

Generalization:

• MARTE::NW.

Associations:

•

Topology: Topology [2], speciﬁes topologies of

the connection between BS and routers, and the

connection between routers and EndDevices.

•

sna: SubNetworkArea [1..*], the network is com-

posed of a set of SNAs.

•

BS: BaseStation[1], the network contains only one

base station.

Attributes:

• None.

Semantics:

•

A network is composed of one Base Station (BS)

and a set of sub network area (SNA) containing

a deﬁnite number of nodes. It is characterized by

its topology which deﬁnes the relations between

nodes.

ENASE 2017 - 12th International Conference on Evaluation of Novel Approaches to Software Engineering

206

5.2 «SubNetworkArea» Stereotype

Generalization:

• None.

Associations:

•

Topology: Topology [2], speciﬁes topologies of

BS and routers, and routers and EndDevices.

•

endpoint: EndPoint [1..4], each SNA contains four

end devices which are connected to the router.

•

router: Router [1], only one router exists in each

SNA.

Attributes:

•

HorizentalArea: NFP_Area, Each SNA has a deﬁ-

nite horizontal area.

•

Level: integer, the SNA is composed of a set of

vertical down levels.

•

NbreNode: integer, each SNA contains a deter-

mine number of nodes

Semantics:

•

A sub network area is the environment that groups

a set of nodes. Each SNA has a determined width

and a set of levels which each level consists of 5

nodes one of each acts a router and others are end

devices.

5.3 «BaseStation» Stereotype

It inherits attributes and implementation of Node class.

Generalization:

• MARTE::GQAM::GaExecHost::Node.

Associations:

• None.

Attributes:

•

CoverageArea: integer, the coverage area of the

base station

Semantics:

•

One base station is considered in the network,

which is ﬁxed and can be replaced by any other

node at run time. It is responsible for network ini-

tiation, determines the ﬁrst router of every SNA,

receives data from nodes and transmits data to the

attached system.

5.4 «Topology» Stereotype

It inherits attributes and implementation of HwCom-

municationResource from GRM MARTE subproﬁle.

Generalization:

•

MARTE:: GRM:: CommunicationMedia:: Hw-

CommunicationResource

Associations:

• None.

Attributes:

•

Entity: Node, describes the function of the node in

the network topology ( parent node, central hub)

Semantics:

•

Two possible topologies can exist in the network.

The tree topology for the connection between BS

and routers, and the star topology for the connec-

tion between router and End Devices.

5.5 «Node» Stereotype

It inherits attributes and implementation of GaExe-

cHost from GQAM MARTE sub proﬁle.

Generalization:

• MARTE::GQAM::GaExecHost

Associations:

•

batteryLevel: BatteryLevel[1..*], indicates the re-

maining energy in each sensor node.

•

rssi: RSSI[1..], indicates the received srenght sig-

nal indicator.

Attributes:

•

Batterylevel: BatteryLevel, the remaining energy

in the battery.

•

RSSIValue: RSSI, received srenght signal indica-

tor.

•

NodeId: integer, the ID of each node in the net-

work.

Semantics:

•

A node represents physical processing device ca-

pable of storing and executing program code. Each

node is deﬁned in the network by an ID. This class

presents an attribute to deﬁne the battery level and

an attribute to calculate the RSSI value. They are

the two major parameters used to achieve the dy-

namic reconﬁguration of the network.

Network Reconﬁguration for Wireless Sensor Networks using UML/MARTE Proﬁle

207

5.6 «BatteryLevel» Stereotype

Generalization:

• MARTE::NFPs::NFP_Power

Associations:

• None.

Attributes:

•

Threshold: NFP_Power, the minimal level of en-

ergy remaining in the node. This attribute enables

the reconﬁguration of the network when the battery

level of a considered node is lower than its value.

Semantics:

•

A BatteryLevel is the indication of the remaining

battery life to support the mote’s operation.

5.7 «RSSI» Stereotype

Generalization:

• MARTE::NFPs::NFP_Power

Associations:

• None.

Attributes:

•

TxRxDistance: integer, the distance between the

sender and receiver nodes.

Semantics:

•

RSSI value presents the power of received signal

which gives information about the strength of the

received signal and it decreases with the increase

in distance between sender and receiver.

6 CASE STUDY

In order to illustrate our proposed approach for the

modeling of network reconﬁguration, we consider, in

case study form, a water distribution network. Water

pipeline monitoring is one of the most interesting ap-

plications in WSNs. The implementation of sensor

network is a great solution for optimal water quality

and quantity control and leakage management in the

pipe. The architecture of the proposed water pipeline

network is described in Figure 3. The water pipeline

network is divided into several SNA that each SNA

is composed of different levels. We present in Fig-

ure 3 only one SNA with three levels. Following our

proposed approach, in the ﬁrst phase, we begin by

deﬁning a state machine representing possible roles of

existing nodes in the network. We deﬁne three states of

the nodes: ExistingNode, EndPointNode and RouterN-

ode. To do the distribution of node’s role in every

Figure 3: Water pipeline network architecture.

Figure 4: The state machine of network reconﬁguration of

existing nodes.

Figure 5: The state machine of the incoming node role re-

conﬁguration.

SNA, transitions are ensured by event triggering. For

example, the switch from ExistingNode to RouterN-

ode occcurs when the node has the lowest RSSI value

and the greater battery level. Existing nodes require

also function reconﬁguration due to the battery level

reduction in order to avoid disconnectivity in the net-

work. For example, the switch from a RouterNode

to EndPointNode occurs when the battery level of the

considered router becomes lower than a certain thresh-

old and the EndPonitNode with the lowest RSSI value

in the same level becomes the RouterNode. Figure 4

shows the state machine of network reconﬁguration of

existing nodes.

Additionally, the bring or the removal of nodes

in/from the coverage area of the network requires the

reconﬁguration of the network and the distribution of

node’s functions. The incoming mote can be Router

or EndPoint depending on the RSSI value. If a node is

removed from the network, the BS is informed about

the missing node and replaces it. If the missing node

was acting as a router then another node from that

level automatically becomes router to maintain the

network’s connectivity at all times. We present in

Figure 5 the state machine describing the bring of the

node.

ENASE 2017 - 12th International Conference on Evaluation of Novel Approaches to Software Engineering

208

7 CONCLUSION

The present paper presented a high level methodol-

ogy for modeling the architectural reconﬁguration in

WSNs systems. We addressed the lack of network

reconﬁguration semantics in the UML/MARTE pro-

ﬁle and we presented a new proﬁle to specify this

reconﬁguration scenario in WSNs system at high ab-

straction levels. We gave an overview of the MARTE

capabilities for the speciﬁcations of reconﬁgurables

WSNs which shown that the present version of the

MARTE standard is limited to the description of node

level based reconﬁguration. We then presented in de-

tails our proposed proﬁle «NW_Reconﬁguration» to

characterize network reconﬁguration features. The

main contribution of the present work is that it sup-

plies the MARTE proﬁle with new semantics to design

network reconﬁgurability in WSNs applications. We

illustrate the practical use of our proposed with a wa-

ter distribution network. We plan in future works to

investigate the veriﬁcation of Non Functional Prop-

erties (NFPs) such as energy consumption and time

constraints. We also intend to integrate the proposed

proﬁle in an MDE-based approach for the development

of low power WSNs.

ACKNOWLEDGMENTS

This work was supported by King Abdulaziz City for

Science and Technology (KACST) and Digital Re-

search Center of Sfax (CRNS).

REFERENCES

Al Ameen, M., Liu, J., and Kwak, K. (2012). Security

and privacy issues in wireless sensor networks for

healthcare applications. Journal of medical systems,

36(1):93–101.

Alam, M. Z., RehmanSheikh, M. B. U., Zeshan, A., et al.

(2014). Topology aware auto-conﬁguration in wireless

sensor network. International Journal of Engineering

Research.

Benatia, M. A., Louis, A., Baudry, D., Mazari, B., and Hami,

A. E. (2014). Wsn’s modeling for a smart building

application. In Energy Conference (ENERGYCON),

2014 IEEE International.

Cherif, S., Trabelsi, C., Meftali, S., and Dekeyser, J. L.

(2011). High level design of adaptive distributed con-

troller for partial dynamic reconﬁguration in fpga. In

Design and Architectures for Signal and Image Pro-

cessing (DASIP), 2011 Conference on.

de Lima, G., e Silva, L., and Neto, P. (2010). Wsn as a

tool for supporting agriculture in the precision irriga-

tion. In Networking and Services (ICNS), 2010 Sixth

International Conference on, pages 137–142.

Ebeid, E., Medina, J., Quaglia, D., and Fummi, F. (2015).

Extensions to the uml proﬁle for marte for distributed

embedded systems. In 2015 Forum on Speciﬁcation

and Design Languages (FDL).

Eronu, E., Misra, S., and Aibinu, M. (2013). Reconﬁguration

approaches in wireless sensor network: Issues and

challenges. In 2013 IEEE International Conference on

Emerging Sustainable Technologies for Power ICT in

a Developing Society (NIGERCON).

Gamatié, A., Le Beux, S., Piel, É., Etien, A., Ben Atitallah,

R., Marquet, P., and Dekeyser, J.-L. (2008). A Model

Driven Design Framework for High Performance Em-

bedded Systems. Technical report, INRIA.

Group, O. O. M. (June 2011). A UML Proﬁle for MARTE:

Modeling and Analysis of Real-Time Embedded sys-

tems, ptc/2011-06-02. Object Management Group.

Krichen, F., Hamid, B., Zalila, B., and Jmaiel, M. (2011).

Towards a model-based approach for reconﬁgurable

dre systems. In Software Architecture. Springer.

Mottola, L. and Picco, G. P. (2011). Programming wireless

sensor networks: Fundamental concepts and state of

the art. ACM Comput. Surv., 43.

Quadri, I. R., Meftali, S., and Dekeyser, J.-L. (2009). A

Model based design ﬂow for Dynamic Reconﬁgurable

FPGAs. International Journal of Reconﬁgurable Com-

puting.

Rajasekaran, C., Jeyabharath, R., and Veena, P. (2014).

Hardware-software reconﬁgurable techniques for wire-

less sensor network. Research Journal of Applied Sci-

ences, Engineering and Technology.

Said, M. B., Kacem, Y. H., Amor, N. B., Kerboeuf, M.,

and Abid, M. (2013). Fine-grain adaptation for real

time embedded systems using uml/marte proﬁle. In

Speciﬁcation Design Languages (FDL), 2013 Forum

on.

Schmidt, D. C. (2006). Guest editor’s introduction: Model-

driven engineering. Computer, 39(2):25–31.

Vidal, J., de Lamotte, F., Gogniat, G., Diguet, J. P., and

Guillet, S. (2011). Dynamic applications on recon-

ﬁgurable systems: From uml model design to fpgas

implementation. In 2011 Design, Automation Test in

Europe.

Vidal, J., de Lamotte, F., Gogniat, G., Diguet, J. P., and

Soulard, P. (2010). Uml design for dynamically re-

conﬁgurable multiprocessor embedded systems. In

2010 Design, Automation Test in Europe Conference

Exhibition (DATE 2010).

Network Reconﬁguration for Wireless Sensor Networks using UML/MARTE Proﬁle

209