Transmission range in VANET (vehicular networks)

[Avaneesh]

Hello,

I am working in the area of clustering in vehicular adhoc networks(vanet). I noticed that ns-3 does not have an implementation of clustering in VANETs, hence I am implementing this as a module. I have noticed that research papers in the area often use Two Ray Ground Propagation model and specify a transmission range (such as 100 meters). My problem is, how are the TxStart TxGain RxGain etc. parameters of wifiChannel converted to range in meters. 

I have attached the conference paper I am following. In Section 4.1 (Clustering Process Scenario), various WAVE configuration parameters are given, which correspond to 60m, 100m and 150m transmission range.

I want to perform simulation on 200m and 300m transmission range. What parameter values should I use? 

The source file corresponding to the attached paper is here. 

Here is the code snippet for transmission parameters:

    // The below set of helpers will help us to put together the wifi NICs we want
    YansWifiChannelHelper wifiChannel = YansWifiChannelHelper::Default ();
    Ptr<YansWifiChannel> channel = wifiChannel.Create ();

    YansWifiPhyHelper wifiPhy =  YansWifiPhyHelper::Default ();
    wifiPhy.SetChannel (channel);
    wifiPhy.SetPcapDataLinkType (YansWifiPhyHelper::DLT_IEEE802_11);
    wifiPhy.Set ("TxPowerStart", DoubleValue(32));
    wifiPhy.Set ("TxPowerEnd", DoubleValue(32));
    wifiPhy.Set ("TxGain", DoubleValue(12));
    wifiPhy.Set ("RxGain", DoubleValue(12));
    wifiPhy.Set ("EnergyDetectionThreshold", DoubleValue(-61.8));
    wifiPhy.Set ("CcaMode1Threshold", DoubleValue(-64.8));




Thanks in advance!

[Tommaso Pecorella]

Hi,

the topic of transmission range in Wi-Fi networks (ad-hoc, infrastructure or valets) has been discussed so many times that I promised to myself to not give any advice to anyone asking again about it.

Search the group and post again if and only if you have really a different problem. Note that using a different propagation model doesn't classify it as a different problem.

T.

[Avaneesh]

Tommaso,

I have searched the group thoroughly, and even read the ns-3 training slides. (including the slide where there is a graph how range varies, no idea how that was generated)

The question has been asked many times, but still proper answers were not given in any post.

First of all, being from computer science background, I do not know what is meant by "TxGain" and "RxGain". Can you explain to me what is transmission gain and reception gain? 

I've googled but these do not seem to be standard terminology.

Thanks

[Tommaso Pecorella]

Hi,

if you work (or you want to work) with wireless communications you should have some wireless communication basic understanding, like antennas, how energy is transmitted, and so on. Otherwise you'll never understand some basic effects like multi path fading, hidden terminals, doppler and so on. I'd strongly suggest to grab a "101" book about wireless communication.

TxGain and RxGain are semi-standard terminologies: Tx stands for Transmission, Rx for Reception. Gain is Gain... 

There are some formulas (see for example https://en.wikipedia.org/wiki/Link_budget) that will help in calculating when the received power falls below a given threshold, but defining exactly what is the radio range is not simple. Meaning, even if you're in radio range, there's always the possibility to have a packet loss.

In other terms, you need to understand at least the basics. Start with the link budget formulas and go deeper on each of its topics.

T.

[Avaneesh]

Thanks a lot for your reply Tommaso. I will check the links which you have provided.

From your reply, I have concluded that in a research paper, while presenting the results of an ns-3 simulation, it will be much better to quote the various simulation parameters (Tx/Rx Power/Gains, PowerLevels, Energy-thresholds, data rate etc.) rather than the transmission range in meters. And the variation needs to be demonstrated in TxPowerStart rather than range in meters. In other words, there needs to be no talk about range in meters anywhere (for a realistic propagation loss model).

Is this a correct conclusion?

[Tommaso Pecorella]

Hi,

it depends on the audience. If your paper is aimed at computer scientists, stick with the range in meters. If you want to be an overachiever, put all the parameters and then a "which corresponds to about x meters".

Note that the "x meters" can be easily found just by doing a simulation and moving two nodes.

Also note that (I'm telling because you're a computer scientist and you could not be aware of some tricks).

- The antenna gain is due to the antenna geometry and its building process. You can't give it unrealistic values and you can't easily change it once you have built your hardware. Sure you can change the antenna, but it's not something you can change easily.

- The Rx power is your goal - you don't change it, you want it :)

- The path loss is due to your channel.

- What you usually change is the Tx power.

- ED and CCA thresholds are (again) parameters that are due to manufacturing. You don't change them, you read them from the hardware data sheets.

Cheers,

T.