Smart Wireless Sensor Networks
model speciﬁes the topology, i.e. the structure of how the nodes are organized, there are different topologies to a WSN such as square, star, ad-hoc, irregular Piedrahita et al. (2
ig. . ardware, pplication ayers and omplete odel Proposal
6.2.2 Middle Layer
he middle layer is responsile to attach a WSN with the needed agents for a speciﬁc application. ence this layer has two agents that perform control and resources manage.
•anager resources gent ( t is a specialized moile agent that taes decisions aout controlling resources of memory and power. t is aware of required charge for an agent performs a tas, and denies or admits to eecute an agent. his is an agent that taes decisions ased on a model eorgeff et al. (. oreoer, it says if a group of tass can e eecuted in eeping with the speciﬁed hardware.
•Capturing Agent of physical variables (CA): It is a mobile agent that is aware of physical variables according to a speciﬁc application. It takes decisions about propagation and transmitting of these variables.
6.2.3 Application Layer
he application layer represents speciﬁc study case or application for which the is going to be deployed. herefore this layer has agents that perform application reuired tasks.
•Coordinator Agent (CoA): It is an agent aware of reuired tasks by a study case so it has a ueue of application tasks. ence it manages organies and negotiates them for being eecuted by a A successfully. Also it takes decisions based on a I model.
•asks Agent (A): It is a reactive agent that performs tasks assigned by a CoA as long as CoA said it had to be.
•eliberative Agent (A): It is a mobile agent that takes decisions based on a I model too. It does not need that a CoA manages organies and negotiates its tasks it does by its own. Accordingly it performs a set of tasks to achieve its own goal or a goal established by a A which it belongs to.
It is a speciﬁc treatment for an application multi agent system due to not all sensor nodes platforms can perform a rational agent i.e. for a simple application there is a group of A with a CoA that manages and coordinates entire system and for a comple application there is a group of A that interact to achieve a global goal.
6.3 Interaction Process
irst of all the CoA(or a A depending of reuired type agents) starts the process for assign ing a task it has the belief that a task needs to be done it has this belief because there is a tasks list related to the application. Its desire consist of ensure that a task is done successfully by a A. hen its ﬁrst intention is to interact with A and to ask task feasibility.
ow A beliefs about its hardware characteristics and charge task and its desire consist to inform if there are enough resources to do the task for this reason its intention is reasoning if charge task processing ﬁts on available resources. It informs true or false.
If A answer is true CoA second intention is to create an instance of a A and assign this task. inally its last intention is to be sure that the task was done then it asks to A if it is done and depending on this answer it starts with another task or the same.
In the case of A multi agent system any A starts the interaction process with agents in the middle layer. A beliefs about its hardware characteristics and charge on a plan (task group). If A conﬁrms available resources the A starts its process otherwise it waits until get an afﬁrmation from A.
aking into account above process we introduce some theoretical formula to determinate global battery discharge (see uation and ) and memory usage (see uation and ) for a time period in the simulation.
B(t) = B(t−) − P(CoA)(MA) − P(TA)L(−)
and P(MA) are the processing of DA and MA agents and L(t−1) is the plan charge. These tasks and plans are negotiated in a speciﬁed order, and constantly repeating.
For Memory usage (M(t)), the formula required to perform or not a task or a plan,
M(t) = M(t−1) − P(CoA)(MA) − P(TA)L(t−1) + P(TA)L(t−)
7. Conclusions and future work
The principles, algorithms and application of Distriuted Artiﬁcial ntelligence can e used to optimie a net ork of distriuted ireless sensors. The MultiAgent ystem approach permits optimiation using rational agents to get this achieement.
t is possile to implement a solution that enales a sensor net ork to ehae as an intelligent multiagent system through the proposed model due to it utilies multiagent systems together ith layered architecture to facilitate intelligence and simulate any , all needed is to kno the ﬁnal application, here the is going to e deploy. Also, a layered architecture can proide modularity and structure for a system. Moreoer, proposed model emphasies aout ho a orks and ho to make it intelligent.
From a perspectie of multiagents, artiﬁcial societies and simulated organiations, a distriuted sensor net ork can e installed in an efﬁcient manner and achiee the proposed o ecties of taking measures of physical ariales y itself ith different types of rational agents that can e reconﬁgured to ﬁt any kind of application and measures, also to ﬁt the most appropriate strategy to achiee requirements of physical ariales monitoring.
Further ork to do is testing model using a real . ome study cases of multiagent systems for speciﬁc applications are required to do a complete testing. A useful tool to use is the
olarium unT emulator. This emulator makes aailale a realistic testing to deelop and test unT deices ithout requiring hard are platform. After this testing ﬁnishes, the model could e performed oer a real of unT deices.
This ork presents the results of the researches carried out y DA (Artiﬁcial ntelligence esearch Deelopment roup) and (cientiﬁc ndustrial nstrumentation esearch roup) at the ational niersity of olomia ampus Medellin, as adance of t o research pro ects cosponsored y DM (esearch Direction of ational niersity of olomia at Medellin ampus) and A (olomian nstitute of cience and Technology) respectiely entitledntelligent yrid ystem Model for Monitoring of hysical ariales using and MultiAgent ystems ith code 11 and Deelopment of a model of intelligent hyrid system for monitoring and remote control of physical ariales using distriuted ireless sensor net orks ith code 1.
heong, . (). Actororiented programming for ireless sensor net orks.
onte, ., ilert, . ichman, . (1). MA and social simulation A suitale commitment, Multi-Agent Systems and Agent-Based Simulation, pringer, pp. 1–.
CRULLER, D., Estrin, D. & Srivastava, M. (2004). Overview of sensor networks, Computer
Davoani, D., art, E. & aeter, . (200). n i neinsire aroa to seke
o tin, Artiﬁcial Immune Systems . 288–2.
EeaLoe, E., aesonso, ., MartineSaa, ., avonMarino, . & ariaaro, . (200).
Si ation saaiit isses in wireess sensor networks, IEEE Communications Magazine 44(): 4.
eoreff, M., e, ., oak, M., a e, M. & oorie, M. (8). e eiefesire intention oe of aen, Intelligent Agents V. Agent Theories, Architectures, and Languages th International orshop, ATAL’ , aris, rance, uly . roceedings, Sriner, . 0–0.
Levis, ., Lee, ., es, M. & Cer, D. (200). OSSM: rate an saae si ation of entire inOS aiations, roceedings o the st international conerence on Em edded netored sensor systems, CM, . .
Moreno, ., eáse, . & Ovae, D. (200). Una roi aión Metooóia ara a Con strión e Moeos e Si aión asaos en e arai a Mtiente, Aances en Sistemas e Inormática 4(2).
O’are, ., O’ra, M. & Mars, D. (200). tono i wireess sensor networks: nteient
iitos sensin, proceeding o AILA , International Congress on Methodologies or Emerging Technologies in Automation, iser, Universit La Saiena, Ro e, ta.
ieraita, ., Montoa, . & Ovae, D. (200). erfor ane Evaation of an nteient
entsase Moe in S wit irrear toooies.
Ro er, . & Mattern, ¡. (2004). e esin sae of wireess sensor networks, IEEE ireless
Communications 11(): ¢4–.
Rsse, S. & orvin, . (200). Artiﬁcial Intelligence A Modern Approach, rentiea, En
Sa, ., ar, M., n, S. & ison, . (2008). Resore anae ent in wireess sensor networks sin oetive inteiene, International Conerence on Intelligent Sensors, Sensor etors and Inormation rocessing, . ISSI , . 42–428.
an, £., an, S. & ian, . (200). Oti ie eo ent strate of oie aents in wireess sensor networks, Intelligent Systems esign and Applications, . ISA’. Sith International Conerence on, o. 2.