Research On Algorithm Of Energy Conservation In WSN

Bring mobile sink into WSN can balance the energy of the network and prolong thelifetime of network.In this paper we will impove the TTDD protocol which is a classicalprotocol in WSNBMS and also we will propose a kind of sink mobile strategy that based onthe residual energy level.After research and analysis of TTDD we find out that in this protocol sink get thetransmit path through the way of reverse finding.In this way, the transmit path is too longand the transmit nodes are troppo. So we propose a new protocol (N-TTDD) to make upthese problems. N-TTDD algorithm fusion the virtual grid thoughts of the former TTDDalgorithm and based on this, taking the source node as the origin of the coordinate system.So the vertex positions of the grid can be coordinated.In this way, we can simplify thecomplex network topology structure.The data transmission process in this protocol can bedivide into three stages: first of all, the the source node will send a notification message toevery forwarding node through the gird when it reactions the data.Sink will flood abroadcast message in the gird limits and the first forwarding node that receive this messagewill become the immediate forwarding node of sink and we need to record itscoordinate.Secendly, in the girds which are between the sink and source node, we need tolateral cmpare and progressively reduce their coordinates. In this way, we can choose thediagonal position of one grid as its new forwording node and these new forwording nodescan make up the new transmit path. So the transmit path can be locality optimized.Then wecan expand the locality optimization based on the grid to the whole network.In this way, thetransmit path can be global optimized.Thirdly, source node send the data to sink through theoptimized transmit path.In this protocol, we also introduce preinstall delay to thenotification message, in this way we can adjust the lifetime of the grid and the coordinatedynamicly and then the extra expenses of the network will be reduced.In consideration of the residual energy of the local nodes and the number of theneighborhood nodes, we propose a kind of sink mobile strategy based on the residualenergy level (MSBREL) in this paper. Sink adopt the walk-stay way in this algorithm, it cantransmit data only when it stay on one postion. In order to reduce the delay of the network,we set that the stay time of sink is much longer than its move time. When we select thepostiton that sink can stay on, first of all, we need to give the corresponding weights tothe residual energy of the local nodes and the number of the neighborhood nodes indifferent network topology structures. Then the node needs to set up a packet whichcontains its residual energy and neighborhood nodes number. After this, the node sends this packet to sink then we can work out this node’s weighted price with this packet and theweights. Then wen will introduce a new data sheet which contains every position’sweighted price and its coordinate.We need to compare the weighted price and the value ofthe top of the stack and the distance of the node and sink and distance of the top of the stackand the sink.After this, we can push the weighted price and the coordinate on two differentstacks.When this comparion and the push into stacks finished, the value of top of thesecond stack is the coordinate of the postion which sink will stay on. What’s more, in orderto reduce the extra expenses of the network which brings from the meaningless mobile ofsink, we propose a new concept: residual energy level. We divide the residual energy of thesensor nodes into different levels. Only if the level of the certain postion which sink stay onis lower than the destination level, the sink can move to the destination postion. In thimethod, this problem can be solved. So, this new move strategy can increase the networklife time.The algorithms proposed in this paper all execute under the NS2simulation.Simulation carried out in a number of scenarios and the results are scientific andvalidtity. We compare the network lifetime, average delay and the network energyconsumption between N-TTDD, DD and TTDD and find out that N-TTDD isadvantageous.We compare he network lifetime, average delay, the network energyconsumption and packets success delivery rates between MSBREL, Static Sink andRandom Move and find out tht the network lifetime of MSBREL is longest and it alsoconsume the least energy. The average delay and the packet success delivery rates ofMSBREL is not better than the Static Sink, but still in the may meet range.