nam能够实现,xgraph能够进行简单的结果统计。昨天做了初步统计结果的东西,我觉得有必要重看论文,然后找出论文的评价因素。
重现HHVBF的实验结果:
路由层agent模拟HH-VBF和VBF,应用层agent模拟source和sink,broadcastMac没有冲突控制手段,1 pkt/10s,range:100m,能量消耗:2w(sending),0.75w(recv),8mw(idle),packet size:50 B,pipe radius:100m,beta:75m。每个节点选择一个目的节点并且向这个节点移动,仿真时间:1000s。
三个评价标准:
(1)success rate: the ratio of the number of packets successfully received by the sink to the number of packets generated by the source
(2)energy cost: total energy consumption of all the nodes in the network
(3)energy tax: the average energy consumption for each successfully received packet
可能会对结果产生的影响因素:
(1)节点密度
success rate vs node density:{x:节点数目;y:成功率}成功率是怎么测量?
随着节点密度的增加,成功率增加,奇怪了,vbf_example_6.tcl为啥中间节点都无法接收到数据呢?
SINK 0 : terminates (send 0, recv 8, cum_delay 59.973402)
SINK(0) : send_id = 1, num_recv = 8
SINK 1 : terminates (send 20, recv 0, cum_delay 0.000000)
SINK 2 : terminates (send 0, recv 0, cum_delay 0.000000)
sink(0)节点接收到8个数据包,所有包均来自sink1(node1),也就是说节点都没有经过中间节点了,直接从sink1到节点sink0。
每5秒中发送节点
stop=100,interval=20.0
node=10,sink1:10,sink0:0
node=20,sink1:10,sink0:0
node=50,sink1:10,sink0:1
node=60,sink1:10,sink0:0
node=70,sink1:10,sink0:0
node=80,sink1:10,sink0:2
node=90,sink1:10,sink0:1
node=100,sink1:10,sink0:
recv=5,cum_delay=30.199520,total_energy=113.335344
3,27.907556,119.044393
1,5.528712,98.358784
为什么recv为0,除了dest节点,其通过DataTable能够获得该节点接收节点的来源。衡量成功率主要是看sink recv/source send
energy cost vs node density:{x:节点数目;y:能量消耗},能量消耗怎么算?
energy tax vs node density:{x:节点数据;y:能量tax}
(2)节点移动性
Success rate vs node speed
Energy cost vs node speed
Energy tax vs node speed