以wifi-example-sim.cc为例说明NS3统计数据模型

利用NS3已有的Trace系统或者Log机制收集记录和统计数据，例如MAC层收发帧数目，网络层以上收发包数目的跟踪与统计，这里选取example/stats/wifi-example-sim.cc为例来很好说明问题：

这个仿真程序是一个简单的实验，包括两个节点，基于AdhocMAC信道模型，包含NS3仿真所需常见模型如节点/网络设备/协议栈和应用进程，这里的应用进程Sender 和Receiver，基于UDP的不可靠连接。

/* -*- Mode:C++; c-file-style:"gnu"; indent-tabs-mode:nil; -*- */
/*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License version 2 as
* published by the Free Software Foundation;
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
*
* Authors: Joe Kopena <tjkopena@cs.drexel.edu>
*
* This program conducts a simple experiment: It places two nodes at a
* parameterized distance apart.  One node generates packets and the
* other node receives.  The stat framework collects data on packet
* loss.  Outside of this program, a control script uses that data to
* produce graphs presenting performance at the varying distances.
* This isn't a typical simulation but is a common "experiment"
* performed in real life and serves as an accessible exemplar for the
* stat framework.  It also gives some intuition on the behavior and
* basic reasonability of the NS-3 WiFi models.
*
* Applications used by this program are in test02-apps.h and
* test02-apps.cc, which should be in the same place as this file.
*
*/

#include <ctime>

#include <sstream>

#include "ns3/core-module.h"
#includ
4000
e "ns3/network-module.h"
#include "ns3/mobility-module.h"
#include "ns3/wifi-module.h"
#include "ns3/internet-module.h"

#include "ns3/stats-module.h"

#include "wifi-example-apps.h"

using namespace ns3;
using namespace std;

NS_LOG_COMPONENT_DEFINE ("WiFiDistanceExperiment");

void TxCallback (Ptr<CounterCalculator<uint32_t> > datac,
std::string path, Ptr<const Packet> packet) {
NS_LOG_INFO ("Sent frame counted in " <<
datac->GetKey ());
datac->Update ();
// end TxCallback
}

//----------------------------------------------------------------------
//-- main
//----------------------------------------------
int main (int argc, char *argv[]) {

double distance = 50.0;
string format ("omnet");

string experiment ("wifi-distance-test");
string strategy ("wifi-default");
string input;
string runID;

{
stringstream sstr;
sstr << "run-" << time (NULL);
runID = sstr.str ();
}

// Set up command line parameters used to control the experiment.
CommandLine cmd;
cmd.AddValue ("distance", "Distance apart to place nodes (in meters).",
distance);
cmd.AddValue ("format", "Format to use for data output.",
format);
cmd.AddValue ("experiment", "Identifier for experiment.",
experiment);
cmd.AddValue ("strategy", "Identifier for strategy.",
strategy);
cmd.AddValue ("run", "Identifier for run.",
runID);
cmd.Parse (argc, argv);

if (format != "omnet" && format != "db") {
NS_LOG_ERROR ("Unknown output format '" << format << "'");
return -1;
}

#ifndef STATS_HAS_SQLITE3
if (format == "db") {
NS_LOG_ERROR ("sqlite support not compiled in.");
return -1;
}
#endif

{
stringstream sstr ("");
sstr << distance;
input = sstr.str ();
}

//------------------------------------------------------------
//-- Create nodes and network stacks
//--------------------------------------------
NS_LOG_INFO ("Creating nodes.");
NodeContainer nodes;
nodes.Create (2);

NS_LOG_INFO ("Installing WiFi and Internet stack.");
WifiHelper wifi;
WifiMacHelper wifiMac;
wifiMac.SetType ("ns3::AdhocWifiMac");
YansWifiPhyHelper wifiPhy = YansWifiPhyHelper::Default ();
YansWifiChannelHelper wifiChannel = YansWifiChannelHelper::Default ();
wifiPhy.SetChannel (wifiChannel.Create ());
NetDeviceContainer nodeDevices = wifi.Install (wifiPhy, wifiMac, nodes);

InternetStackHelper internet;
internet.Install (nodes);
Ipv4AddressHelper ipAddrs;
ipAddrs.SetBase ("192.168.0.0", "255.255.255.0");
ipAddrs.Assign (nodeDevices);

//------------------------------------------------------------
//-- Setup physical layout
//--------------------------------------------
NS_LOG_INFO ("Installing static mobility; distance " << distance << " .");
MobilityHelper mobility;
Ptr<ListPositionAllocator> positionAlloc =
CreateObject<ListPositionAllocator>();
positionAlloc->Add (Vector (0.0, 0.0, 0.0));
positionAlloc->Add (Vector (0.0, distance, 0.0));
mobility.SetPositionAllocator (positionAlloc);
mobility.Install (nodes);

//------------------------------------------------------------
//-- Create a custom traffic source and sink
//--------------------------------------------
NS_LOG_INFO ("Create traffic source & sink.");
Ptr<Node> appSource = NodeList::GetNode (0);
Ptr<Sender> sender = CreateObject<Sender>();
appSource->AddApplication (sender);
sender->SetStartTime (Seconds (1));

Ptr<Node> appSink = NodeList::GetNode (1);
Ptr<Receiver> receiver = CreateObject<Receiver>();
appSink->AddApplication (receiver);
receiver->SetStartTime (Seconds (0));

Config::Set ("/NodeList/*/ApplicationList/*/$Sender/Destination",
Ipv4AddressValue ("192.168.0.2"));

//------------------------------------------------------------
//-- Setup stats and data collection
//--------------------------------------------

// Create a DataCollector object to hold information about this run.
DataCollector data;
data.DescribeRun (experiment,
strategy,
input,
runID);

// Add any information we wish to record about this run.
data.AddMetadata ("author", "tjkopena");

// Create a counter to track how many frames are generated.  Updates
// are triggered by the trace signal generated by the WiFi MAC model
// object.  Here we connect the counter to the signal via the simple
// TxCallback() glue function defined above.
Ptr<CounterCalculator<uint32_t> > totalTx =
CreateObject<CounterCalculator<uint32_t> >();
totalTx->SetKey ("wifi-tx-frames");
totalTx->SetContext ("node[0]");
Config::Connect ("/NodeList/0/DeviceList/*/$ns3::WifiNetDevice/Mac/MacTx",
MakeBoundCallback (&TxCallback, totalTx));
data.AddDataCalculator (totalTx);

// This is similar, but creates a counter to track how many frames
// are received.  Instead of our own glue function, this uses a
// method of an adapter class to connect a counter directly to the
// trace signal generated by the WiFi MAC.
Ptr<PacketCounterCalculator> totalRx =
CreateObject<PacketCounterCalculator>();
totalRx->SetKey ("wifi-rx-frames");
totalRx->SetContext ("node[1]");
Config::Connect ("/NodeList/1/DeviceList/*/$ns3::WifiNetDevice/Mac/MacRx",
MakeCallback (&PacketCounterCalculator::PacketUpdate,
totalRx));
data.AddDataCalculator (totalRx);

// This counter tracks how many packets---as opposed to frames---are
// generated.  This is connected directly to a trace signal provided
// by our Sender class.
Ptr<PacketCounterCalculator> appTx =
CreateObject<PacketCounterCalculator>();
appTx->SetKey ("sender-tx-packets");
appTx->SetContext ("node[0]");
Config::Connect ("/NodeList/0/ApplicationList/*/$Sender/Tx",
MakeCallback (&PacketCounterCalculator::PacketUpdate,
appTx));
data.AddDataCalculator (appTx);

// Here a counter for received packets is directly manipulated by
// one of the custom objects in our simulation, the Receiver
// Application.  The Receiver object is given a pointer to the
// counter and calls its Update() method whenever a packet arrives.
Ptr<CounterCalculator<> > appRx =
CreateObject<CounterCalculator<> >();
appRx->SetKey ("receiver-rx-packets");
appRx->SetContext ("node[1]");
receiver->SetCounter (appRx);
data.AddDataCalculator (appRx);

/**
* Just to show this is here...
Ptr<MinMaxAvgTotalCalculator<uint32_t> > test =
CreateObject<MinMaxAvgTotalCalculator<uint32_t> >();
test->SetKey("test-dc");
data.AddDataCalculator(test);

test->Update(4);
test->Update(8);
test->Update(24);
test->Update(12);
**/

// This DataCalculator connects directly to the transmit trace
// provided by our Sender Application.  It records some basic
// statistics about the sizes of the packets received (min, max,
// avg, total # bytes), although in this scenaro they're fixed.
Ptr<PacketSizeMinMaxAvgTotalCalculator> appTxPkts =
CreateObject<PacketSizeMinMaxAvgTotalCalculator>();
appTxPkts->SetKey ("tx-pkt-size");
appTxPkts->SetContext ("node[0]");
Config::Connect ("/NodeList/0/ApplicationList/*/$Sender/Tx",
MakeCallback
(&PacketSizeMinMaxAvgTotalCalculator::PacketUpdate,
appTxPkts));
data.AddDataCalculator (appTxPkts);

// Here we directly manipulate another DataCollector tracking min,
// max, total, and average propagation delays.  Check out the Sender
// and Receiver classes to see how packets are tagged with
// timestamps to do this.
Ptr<TimeMinMaxAvgTotalCalculator> delayStat =
CreateObject<TimeMinMaxAvgTotalCalculator>();
delayStat->SetKey ("delay");
delayStat->SetContext (".");
receiver->SetDelayTracker (delayStat);
data.AddDataCalculator (delayStat);

//------------------------------------------------------------
//-- Run the simulation
//--------------------------------------------
NS_LOG_INFO ("Run Simulation.");
Simulator::Run ();

//------------------------------------------------------------
//-- Generate statistics output.
//--------------------------------------------

// Pick an output writer based in the requested format.
Ptr<DataOutputInterface> output = 0;
if (format == "omnet") {
NS_LOG_INFO ("Creating omnet formatted data output.");
output = CreateObject<OmnetDataOutput>();
} else if (format == "db") {
#ifdef STATS_HAS_SQLITE3
NS_LOG_INFO ("Creating sqlite formatted data output.");
output = CreateObject<SqliteDataOutput>();
#endif
} else {
NS_LOG_ERROR ("Unknown output format " << format);
}

// Finally, have that writer interrogate the DataCollector and save
// the results.
if (output != 0)
output->Output (data);

// Free any memory here at the end of this example.
Simulator::Destroy ();

// end main
}

一 给定本次仿真参数distance，format，experiment，strategy，runID在初始化的同时也可以通过命令行改变，这些参数用于从多次实验中快速区分和组合数据。

二 创建节点和网络模型

三 安装协议栈，并分配IP

四 设置移动模型，这里为静止，并给定初始位置

五 安装应用，这里安装Sender / Receiver，自定义的见examples/stats/wifi-example-apps.h|cc

六 数据统计与收集，这是本文重点，下面具体分析。

这里创建DataCollector对象来存储运行信息，并通过Trace机制记录收发端帧和分组传输情况。

1 记录发端帧传输（基WIFI MAC对界）

通过CounterCalculator(src/stats/model/basic-data-calculators.h )类实现计数，利用Trace机制，当节点0上wifiNetDevice/Mac/MacTx变化（source），通过Config::Connect关联，定义的TxCallback作为sink函数调用，导致CounterCalculator::update调用即m_count++从而起到计数功能;

2 记录收端帧传输（基WIFI MAC对界）

类似情况1，虽然这里的sink函数是PacketConterCalculator::PacketUpdate（src/network/utils/packet-data-calculators.cc），但是该函数仍然是通过CounterCalculator::update实现计数，即利用Trace机制，当节点1上wifiNetDevice/Mac/MacRx变化（source），通过Config::Connect关联;

3 记录发端分组传输

也是通过PacketConterCalculator::PacketUpdate实现计数，利用Trace机制，当节点0上/Application/*/$Sender/Tx变化（source），通过通过Config::Connect关联，定义的PacketConterCalculator::PacketUpdate作为sink函数调用;

4 记录收端分组接收

由于收端应用Receiver没有定义traced source，故这里没有采用Trace机制，而是直接利用Receiver:;SetCounter直接操作，通过SetCounter显示类型转换，j将appRx赋值给Receiver内部计数器，从而实现计数

以上均是通过PacketConterCalculator(src/network/utils/packet-data-calculators.cc)或者CounterCalculator(src/stats/model/basic-data-calculators.h )实现传输单元的计数，\下一个将通过引入PacketSizeMinMaxAvgTotalCalculator (src/network/utils/packet-data-calculators.h|cc)和MinMaxAvgTotalCalculator(src/stats/model/basic-data-calculators.h)实现单元内大小的记录。

5 记录发端分组大小

这里采用Trace机制，节点0上/Application/*/$Sender/Tx变化（source），通过通过Config::Connect关联，定义的PacketSizeMinMaxAvgTotalCalculator::PacketUpdate作为sink函数调用，从而MinMaxAvgTotalCalculator::Update实现大小的记录。

6 记录端到端产生分组时的延迟

类似情况4，不采用Trace机制，直接利用Receiver:;SetDelayTracker记录传世时延最值/平均值等

七 运行程序命令

八 统计结果输出

对于输出要么OMNet++(纯文本输出格式)要么SQLite(数据库格式输出)，这取决于程序头部定义的参数format，并最终DataCollector对象进行存储。

九 控制脚本实现最后运行

通过 一个简单的控制脚本实现该仿真程序在不同距离下大量重复（作为输入）实验后运行画图。可参考example/stats/wifi-example-db.sh（以后自己写多个不同输入下重复仿真项目时可参考这个）。该运行脚本每次都是基于一个不同的距离作为输入，收集每次仿真结果到SQLite数据库，其中对于每个距离输入，进行5次重复实验以减小波动。全部仿真完成只需几十秒，在完成存储到数据库后，可通过SQLite命令行进行SQL查询。并调用 wifi-example.gnuplot画图

进入该目录



产生data.db数据库，wifi-default.data和wifi-default.eps图



图是对应距离下的丢包率以表征WiFi模型性能。