Một giao thức định tuyến dựa trên các thông số năng lượng trong mạng đặc biệt cảm biến không dây

Journal of Science & Technology 101 (2014) 122-126 A Routing Protocol Based on Energy Parameter in Wireless Sensor Adhoc Network Trung Dung Nguyen', Van Due Nguyen Hanoi University ofScience and Technology, No Dai Co Viet Str., Ha Noi, Viet Nam Received June 10 2013; accepted April 22, 2014 Abstract Wireless sensor adhoc networt< is a topic that is discussed more and more in conferences today In this paper, we propose a new routing protocol in wireless sensor adhoc networt< to use energy in the network efficiently That is called muting dual cntenon, RDC for short Our protocol is based on AODV [6] routing protocol with parameters of path selection are both hopcount and nodes' energy Simulation results stiow that the lifetimes and throughputs in our algonthm are better than that using AODV protocol Keywords: Adhoc, Sensor, Routing, energy efficiency, AODV, RDC WSAN I Introdution There are many new models and new solutions that were proposed to improve the efficiency of the network [1-5] A wireless sensor adhoc network (WSAN) [7,8] consists of a large number of sensor nodes that may be deployed randomly and densely They are capable of sensing many types of information from the envhonment, including temperature, light, humidity, radiation, geological features, seismic vibrations, specific types of computer data and more For this reason, WSAN deployment is very suitable for forestry monitoring application such as forest fire monitoring, humidity parameter gathering and so on However, there are many problems with WSAN for example energy, security, etc and energy limitation is one of its big issues Therefore energy saving becomes a major problem of WSAN network to prolong the lifetime of network The network lifetime is amount of time from the starting operation of network until the stopping operation of the first node in the network In this paper we propose a new routing protocol to solve this problem, which balances the energy between the nodes In the entire network, nodes have different energy consumption Nodes with overloads of data flows will stop working sooner than other nodes When one node stops working, it affects the stability of network So besides using energy efficiency in each node we have to balance the remaining energy in all nodes in order to prevent any node from stopping working too early AODV is a popular routing protocol It uses hopcount in selecting the best path To prolong the lifetime of network, we add ' Corresponding Author Tel: (+U) 913534544 Email; dung.nguyentrungl@hust.vn energy criterion into the route selection process of AODV We put node's energy information into route request (RREQ) and route reply (RREP) messages Nodes that receive these messages will have information about hopcount and energy to select the best path The article is divided into parts Part infroduces to wireless sensor adhoc network, its important issue and proposes an improved routing protocol Part II is an overview of AODV routing protocol The next part is detail of the new protocol suggested above Part IV is simulation results and evaluations The final part discusses about conclusion and works future, Overview of aodv routing protocol Adhoc On-Demand Distance Vector (AODV) [6] is a popular routing protocol in mobile adhoc network It finds a route on demand The operation of AODV is described below Flowchart in Fig I describes the operation at source node Source node needs to find the path lo the destination node, it sends Route Request message (RREQ for short) to all its neighbors and sets timeout to wait Route Reply message (RREP) correlatively The RREP message is a message that includes the path to the destination sent by intermediate node or destination node If source node receives the RREP message with path information, routing process will finish In case source node does not receive the RREP message and the timeout finishes, it continues sending other RREQ, setting new timeout and waiting Journal ofScience & Technology 101 (2014) I22-1Z6 Need to find the path to the i Receive KREQ messages age Select llie best patli witli the shortest hopeount Set timeout for RKEQ Send RREP message to r^ Fig The operation at destination node - AODV In the best path selection step, AODV uses hopcount parameter only So nodes in the center of networks are often selected to fransmit data These nodes are out of energy soon and affect network operation As a result, the entire network will stop working soon To solve this problem we propose new routing protocol descnbed m detail in next session Routing dual criterion protocol (RDC) Fig, The operation at intermediate nod A.ODV At intermediate node, it will k up routmg table first when receiving RREQ n g it has path to the destination nod a d mformation is better than source node' n send RREP message with path information Then the routing process will finish In other case, intermediate node forwards RREQ message to all its neighbors to find new path to the destination node The detail of algonthm is descnbed in flowchart in Fig In Fig, 3, destination node receives RREQ messages from different ways then selects the path with the shortest hopcount and sends back to source node via RREP message After that the routing process finishes As mentioned above, AODV uses flooding process and hopcount criterion for finding the best path From RREQ and RREP messages, nodes will have routing mformation If routing information is new, nodes will add new path in to routing table, otherwise they will update path only with existed routing information in the routmg table In these routing protocols, path selection criterion is hopcount parameter It ensures the shortest route will be selected for saving energy of whole network In this method, nodes in the center or nodes with low energy will stop working rapidly and affect the lifetime of network We propose a routmg protocol called Routing Dual Criterion Protocol (RDC) with two criteria namely hopcount and energy In our protocol, route will be selected by hopcount and nodes' energy So routes with low energy nodes are not selected Thereby it increases the lifetime and throughput of the network The energy in this selection is mm energy of nodes that routing messages go across rq_min_energy field is added mto Route Request RREQ message and rp_energy field is added into Route Reply RREP message These fields are initially set to infinity at the first sending node and are updated when message goes across each node Journal ofScience & Technology 101 (2U14) 122-126 ãHi â -HD đ -31] đ 33 â @ Fig Example of rq_min_energy and rp_energy fields In Fig 4, value of rqminenergy field m RREQ message or rpenergy field in RREP message is initiated equal infinity Node S sends message to node A Node A compares its energy at 12 with value of rq_min_energy/ rp_energy field It means node A has lower energy than value of this field setting equal 12 Then node A sends message to node B to compare and set this field equal And this message is sent to node C m order to compare with its energy Energy of node C at 11 is higher than value of this field So value of this field is not changed By repeating such manipulation we have the lowest energy of nodes in the route that message goes across Fig, Route selection process in RDC example Receive RREQ or RREP messages M=hopcouM + l/min_enerey The combination of hopcount and n energy is presented by formula below: M = a: * hopcount -F y * (1) M is called route mettic Route with lowest M value will be selected In the formula, x and y are weight In your network if hopcount is more important than energy, x will be higher than y and inverse In our paper we choose Y and y are the same value, equal I, The formula is calculated as M= hopcount-\ (2) _energy We have an example as below Add tbe best path into the routing table r^ Fig The best path selection process when usmg RDC routing protocol For example in Fig 5, we need to find a route from node S to node D, We have three routes from S to D Route 1; S-A-B-C-D has mm energy 5, hopcount and A/equal (4 + 1/5), Route S-E-F-GD has energy 9, hopcount and M equal (4 -i1/9), Route 3: S-H-I-K-N-D has mm energy 19, hopcount and M equal (5 + 1/19) If route is selected by hopcount, Route and route will be selected Node B stops working rapidly as well If RDC IS used, route will be selected because of lowest M value So node B continues working and does not affect network Network lifetime is increased The detail of RDC protocol operation is the same as AODV protocol expect the best path selection, AODV and RDC select the best paths based on hopcount and A/value respectively Fig Network lifetime Journal ofScience & Technology 101 (2014) 122-126 S^VWWv -AODV - AODV I-RDC I-RDC Numbei of node (nodes) Numbn- of node (nodes) Fig, Packet Delivery Ratio The detail of the best path selection process in RDC routing protocol is presented in Fig 6, Node receives a RREQ or RREP message It calculates M value as formula (2), And then it chooses a path with lowest M value and adds the best path into routing table Simulation result and evaluation We use NS2 application to simulate and evaluate the model The AODV model used has been published on the NS2 community website [9],[10] and has been verified by many users In our simulation, network lifetime, packet delivery ratio and throughput of new routing protocols RDC are compared with the original routing protocol AODV In the simulation, we create 16 topologies with 45 to 120 nodes m an area of 500m x 500m The detail parameters are described as below Simulation time: 200 second Transmit power 0281Sw Fig Throughput network when usmg AODV and RDC routing protocols The lines with rectangle and lozenge are network hfetmies when using RDC and AODV respectively As shown below, lifetime in RDC use is higher than in AODV use Packet delivery ratio (PBR) is ratio between amount of destination transmitted messages and amount of sent messages Fig, shows the packet delivery ratio when usmg AODV and RDC, The lines with rectangle and lozenge are packet delivery ratios in RDC and AODV use respectively The diagram shows that the PDR using RDC is a bit better than that using AODV, It is not much different between the two algorithms Throughput is total data that is transmitted in amount of time Fig shows the throughput of network when usmg AODV and RDC The lines with rectangle and lozenge are the network throughputs when using RDC and AODV respectively The diagram shows that throughput m RDC is higher than in AODV, Inihal energy, randomly from 80 to 120 joules Conclusion Idle power consumption: I w In this paper we propose new routing protocol called RDC The algorithms are better improved from AODV by adding energy as a criterion It changes the route selection method and balances energy between nodes Therefore it prolongs the lifetime of network and improves the quality of connections Transmission power: 3w Receive power- w Packet size: 512 bytes We have results as below Network lifetime is amount of time from starting operation until the first node out of energy (stops working) Fig shows the lifetime of the The result of simulation shows that network lifetime, throughput and packet delivery ratio using RDC is better than that using AODV, Journal ofScience & Technology 101 (2014) 122-126 References [1] Nguyen Thanh Tung, Gregory K Egan, Bill LloydSmith, Brett Pentland Energy based routing techniques in Mobile Ad hoc and Sensor Networks 2005 13th IEEE Intemalionai Conference on Communications, vol [2] Duy Ngoc Pham, Van Due Nguyen, Van Tien Pham, Ngoc Tuan Nguyen, Xuan Bac Do Trung Dung Nguyen, Claus Kuperschmidt, Thomas Kaiser An Expendmg Ring Search Algorithm For Mobile Adhoc Networks 2010 ATC Infemaiional Conference, pp 39-44 [3] Zheng Wang, Hamid Sadjadpour, Jose Joaquin Garcia-Luna-Aceves The capacity and energy efficiency of wireless adhoc networks with multipacket reception MobiHoc '08 Proceeding of the 9"" ACM international symposium on Mobile adhoc networking and compuUng, (2008) 179-188 [4] Mahfoudh, S Minet, P, Survey of energy efficienl strategies in wireless ad hoc and sensor networks, Networking, 2008 ICN 2008, Seventh International Conference, (2008) 1-7 [5] Ajay Shah, Hitesh Gupta, Mukesh Baghel, Energyefficient rouUng protocol for mobile adhoc networks, ISSN-2248-9622, (2012) 1342-1346 [6] C Perkins, E Beldmg-Royer, and S Das Ad hoc OnDemand Distance Vector (AODV) Routtng RFC 3561 (Ex-perimental), July 2003[7] Stefano Basagni, Marco Conti, Silvia Giordano, Ivan Slojmenovic, Mobile Ad Hoc Networking, Copyright 2004 by the Institute of Electrical and Electronics Engineers, [8] David Culler, Overview of Sensor Networks, University of Cahfomia, Berkeley DeborahEstrin, Mani Snvastav, University of California, Los Angeie, 2004[9] Kevin Fall Kannan Varadhan, "The ns Manual", 2008 [10] http//www.isi edu/nsnam/ns/