سونی هندی‌کم 4K خود را با بزرگنمایی 20 برابر معرفی کرد
کمپانی Ricoh دوربین 360 درجه جدید Theta S را در ایفا معرفی کرد
بررسی اسپیکر Z623 لاجیتک
اینتل تا سال ۲۰۱۹ حافظه‌ SSD با ظرفیت ۱۰۰ ترابایت روانه ...

بکار گیری هوش مصنوعی در بهبود شبکه های حسگر بی سیم قسمت دوم

3. Artificial Intelligence and Multi-Agent Systems
€lassical ‚rtificial  ntelligence aimed at emulating within computers the intellectual and interaction abilities of a human being. he modern approach to ‚rtificial  ntelligence (‚ ) is centered around the concept of a rational agent. ‚n agent is anything that can perceive its environment through sensors and act upon that environment through actuators (ƒussell „ orving, …††‡).‚n agent that always tries to optimiˆe an appropriate performance measure is called a rational agent. Such a definition of a rational agent is fairly general and can include human agents (having eyes as sensors, hands as actuators), robotic agents (having cameras as sensors, wheels as actuators), or software agents (having a graphical user interface as sensor and as actuator). ‰rom this perspective, ‚  can be regarded as the study of the principles and design of artificial rational agents.
However, agents are seldom stand-alone systems.  n many situations they coeŠist and interact with other agents in several different ways. ‹Šamples include intelligent eb software agents, soccer playing robots, e-commerce negotiating agents, computer vision dedicated agents, and many more. Such a system that consists of a group of agents that can potentially interact with each other is called a Multi-‚gent Systems (M‚S), and the corresponding subfield of ‚  that deals with principles and design of multi-agent systems is called Œistributed ‚  (Œ‚ ).

4. Wireless Sensor Networks and Artificial Intelligence
‚n intelligent sensor is one that modifies its internal behavior to optimiˆe its ability to collect data from the physical world and communicates it in a responsive manner, to a base station or to a host system. he functionality of intelligent sensor includes: self-calibration, self-validation, and compensation. he self-calibration means that the sensor can monitor the measuring condition to decide whether a new calibration is needed or not. Self-validation applies mathematical modeling error propagation and error isolation or nowledge-based techniŽues. he self-compensation maes use of compensation methods to achieve a high accuracy. he types of artificial intelligence techniŽues widely used in industries are: ‚rtificial eural etwor (‚), ‰uˆˆy ‘ogic and euro-‰uˆˆy.  ntelligent sensor structures embedded in ireless Sensor etwors result in wireless intelligent sensors. he use of ‚rtificial intelligence techniŽues plays a ey role in building intelligent sensor structures. Main research issues of the Ss are focused on the coverage, connectivity networ lifetime, and data fi-delity.  n the recent years, there has been an increasing interest in the area of the ‚rtificial  ntelligence and Œistributed ‚rtificial  ntelligence and their methods for solving Ss constrains, create new algorithms and new applications for Ss. ƒesource management is an essential ingredient of a middleware solution for S. ƒesource management includes initial sensor-selection and tas allocation as well as runtime adaptation of allocated tas’resources. he parameters to be optimiˆed include energy, bandwidth, and networ lifetime.  n this par

ticular case Distributed Independent Reinforcement Learning proposed the use of collective intelligence in resource management within WSNs (Shah et al., 2. inall, intelligent net wor ing and collaborative sstems are also proposed as components for WSNs’ enhancement.
5. Multi-Agent Based Simulation
€‚ƒS refers to the simulation aim at modeling the behavior of agents in order to anal„e their interactions and conse…uences of their decision ma ing process. †ence, a global result is closel determined b agents’ interactions. In practice, €‚ƒS models are used to repre sent and understand social sstems (‡onte et al., ˆ‰‰, moreover to evaluate new strategies of improvement and politics on different  ind of sstems. Due to €‚ƒS is a recentl area, there are actuall few techni…ues and tools for its development. In fact, some contributions come from sstem simulation, software engineering and agentoriented software engineering (‚ŠS‹. acing this constrain, a methodolog was proposed b ŒIDI‚ research group from National Žniversit of ‡olombia, which defines several stages and artifacts for ever phase of a software lifeccle (€oreno et al., 2‰. ’his methodolog allows the representation of main characteristics of the distributed sstem, including  e aspects such as organi„ation, reasoning, communication, and coordination mechanism, among others. ’he main function of WSN simulators is to emulate a WSN operation and simulate entire characteristics of hard ware for each node in simulated WSN, instead of providing strategies to do a deploment. ’he fundamental idea is to propose a model that enables a highl distributed sensor networ  to behave intelligentl as a multiagent sstem. It is important to note that most simulators are used to simulate a specific sstem, be a €‚S or a WSN, but not both of them. ƒesides, it is needed to identif the relationships e“isting between agents and sensor nodes for getting intelligence from the multiagent sstem and monitoring from the WSN. rom WSNs’ point of view, €‚ƒS provides understanding on WSNŠs performance and networ  autonomous capabilities when acting as an agents societ. In this case, agents collaborate together to save and improve resources within the WSN. inall, €‚ƒS can highl contribute to define de ploment strategies and operation politics related to the simulated application.
6. Multi-agent Model proposal
€odel proposal is a €ulti‚gent hbrid model to simulate the deploment of software agents over an WSN, this is done b a laered architecture that utili„es deterministic models of hardware with agent based intelligence, in order to evaluate different strategies, such as dif ferent agents for a specific application. It utili„es mobile agents to control networ  resources and facilitate intelligence. In order to get this, it is used principal deterministic models for WSN performing, such as, protocol model, which comprises all the communication protocols and their operation usuall depends on the state of the phsical platform of nodes, phsical model, which represents the underling hardware and measurement devices, media model, which lin s the node to the •real world• through a radio channel and one or more phsical channels, batter model that is responsible for chec ing if the node has e“hausted its batter through computing power consumption of the different components, among others (‹gea Lope„ et al., 2–. €oreover, it is added the topolog and phsical variables according to the application that is going to be simulated. ’hen, it is used software agents to perform all tas s re…uired b the application stud case.

6.1 Simulation Models for WSN
Present simulation models try to represent how a WSN works. For example, Egea-Lopez at al., in Egea-Lopez et al. ( hae proposed a general simulation model taking into aount urrent omponents o  a WSN simulator. ­ene, there are seeral deterministi models to represent hardware, enironment, power, radio hannels, among others. €hese models are use ul in the way o  knowing a‚out how a WSN per orms in a real li e ‚ut they do not o  er the potential o  ealuating di  erent strategies o  deployment, moreoer, the simulation nodes num‚er is really  ar o  a real network, due to sala‚ility is a  eted ‚y all reƒuired proessing to simulate omplete hardware.
Later, a new propose is presented ‚y „heong in „heong (…. Some strengths o  this work are the use o  di  erent simulation tools whose are already defined  or WSN Leis et al. (‡, and it permits a direted implementation  rom simulation. ­oweer, „heong proposes a programming paradigm ‚ased on ators, whose are a onept ‚etween o‚ˆets and agents. ‰tors are o‚ˆets with data flow  or ommuniation, ‚ut they are not aware o  its enironment neither a‚le to take deisions  or ating.
‰nother approah is presented ‚y Wang and ‹iang in Wang et al. (, where is presented a strategy to ontrol and optimize resoures in a WSN through mo‚ile agents. Œptimization o  resoures suh as, power, proessing and memory o  deies is done, ‚ut it is not defined how deies and agents are related  or getting this optimization.
6.2 Model Proposal
Žt is proposed a ‘ulti-‰gent hy‚rid model to simulate the deployment o  so tware agents oer any WSN, this is done ‚y a layered arhiteture that uses deterministi models o  hardware with agent ‚ased intelligene, in order to ealuate di  erent strategies, suh as di  erent agents  or a speifi appliation.
We aim to utilize mo‚ile agents to ontrol network resoures and  ailitate intelligene. Žn order to get this, it is used the prinipal deterministi models speified ‚y Egea-Lopez et al. (, these models set  eatures, suh as, plat orm o  nodes, power onsumption, radio hannel and media. ‘oreoer, it is added the topology and physial aria‚les aording to the appliation that is going to ‚e simulated. Finally, it is used so tware agents to per orm all tasks reƒuired ‚y the appliation study ase. ’elow is presented three di  erent layers that let to per orm intelligene through agents oer a WSN.
6.2.1 Hardware Layer
€he hardware layer is responsi‚le to spei y all omponents that are related to harateristis proided ‚y hardware and the enironment where network is going to ‚e deployed. ‘ost models o  this layer are already defined ‚y the present WSN simulators. ’elow it is introdued some models that spei y these omponents.
•Node ‘odel” €his model has ‚een speified ‚e ore ‚y Egea-Lopez et al. (, where a node is diided ‚y protools, hardware and media. Protools operation depends on hardware speifiations and omprises all ommuniations protools o  a node. ­ardware represents the underlying plat orm and measurement deies. ‰nd media, links the node to the ¸Sreal world ˇ