Mobile Ad Hoc Networks - Science topic

Designing a mathematical model for the Ad hoc On-Demand Distance Vector (AODV) routing protocol involves formalizing the behavior, performance metrics, and operational characteristics of the protocol. Below are steps on how to approach the design of such a model:

Step 1: Define the Network Environment

Step 2: Understand AODV Operation

AODV can be understood through its key operational phases:

Step 3: Establish Key Performance Metrics

Define the metrics you are interested in analyzing, including:

Step 4: Mathematical Formulation

Trd=TRREQ+TRREPTrd​=TRREQ​+TRREP​

Te2e=∑i=1D(LR+Tpropagation+Tqueue)Te2e​=i=1∑D​(RL​+Tpropagation​+Tqueue​)

Where TpropagationTpropagation​ is the propagation delay, and TqueueTqueue​ is the queuing delay at each node.

Th=PDR⋅LTtotalTh=Ttotal​PDR⋅L​

Where TtotalTtotal​ is the total time for communication.

PDR=NdeliveredNsentPDR=Nsent​Ndelivered​​

Where NdeliveredNdelivered​ is the number of packets successfully delivered and NsentNsent​ is the total packets sent.

O=NcontrolNdataO=Ndata​Ncontrol​​

E=NRREQ⋅Etx+NRREP⋅Erx+NRERR⋅EtxE=NRREQ​⋅Etx​+NRREP​⋅Erx​+NRERR​⋅Etx​

Step 5: Simulate and Validate

Step 6: Refine the Model

Based on validation results, refine your model to improve accuracy, considering factors like dynamic changes in network topology, varying traffic patterns, or more realistic mobility models.

I would recommend the following paper:

'Dynamic Time Slot Allocation for Mobile Ad Hoc Networks using Directional Antennas' Isaac Frank Agidi

I think selective forwarding attacks existed only for RPL, MANET, and Wireless Sensor Network

Thanks so much for your answer Ritakshi Negi

There are projects available in https://tetcos.com/file-exchange.html covering IDS

I do not see a reason why their layered architecture would be affected. The mobility and capabilities of nodes will certainly be affected, mainly affecting the details of the physical (PHY) layer. This in turn may affect the applicable algorithms and protocols at the link and network layer. I do not think any higher layers should be affected.

Hassam Mughal Great

Hi.,

Is it illegal to copy source code..

Some sites use code released under an open-source license; those sites may be copied. However, if the code is NOT released under open-source, it is definitely illegal to take it and use it directly for your own site. This is a copyright violation, no questions about it..

For software this means it is illegal to copy or distribute software, or its documentation, without the permission of the copyright holder. If you have a legal copy of software you are allowed to make a single archival copy of the software for backup purposes.

I use ns 2.35

what about that?

Dear Jateni Halake

Well there could be multiple possibility for that for example ,

1- you may assigned some selected trusted nodes only to participate in route discover

2- clustering with cluster head can also reduced the route discovery braodcast packets

etc

I have several paper related to secure routing for MANET in highly cited journals , look at my RG profile to view the papers

Dr Adnan Nadeem

http://adnan-nadeem.com/

Dear Abhishek Bajpai

You may use castalia package of OMENT simulator has support of power management of the sensors used in AdHoc and Sensor Networks.

Dr Adnan Nadeem

http://adnan-nadeem.com/

I dont know that much about OPNET, but if you have NetSim, I can explain how to do set-up traffic, and do the 2-hops, in a few simple steps

which simulator is best for simulating VANET with 5G communication capabilities ? NS2 or NS3??

I agree with Wahyu Pamungkas, that no single simulator is best for all the use-cases in VAENTS. However, you opt to choose traffic and network simulators simultaneously for most of the parameters. You have to choose traffic simulator first hand to get mobility traces that you can import further to perform network simulations for several parameters. VanetmobiSIM, SUMU, nctuNS are to name the few for first catagory. Similarly, Opnet, Omnet++ and Matlab are the good examples of the later categories. I hope, it helps

For me,

Integration of MANETs node with UAVs.

MANET is pretty well defined as an Internet Protocol scheme. A MANET is built at layer 3, in short. You can go to the IETF site and find multiple RFCs written about MANET, including a version of OSPF routing protocol for MANET, and other routing protocols such as DSDV, OADV, and DSR developed for MANET. Layer 3 schemes, all of them. They work with IP addresses.

Wireless ad-hoc networking is a more generic term. It can apply to layer 2 schemes, such as the use of IEEE 802.11 (WiFi) in so-called "ad hoc" mode, as opposed to the more typical use of WiFi in "infrastructure " mode. For example, for vehicular ad-hoc networks, or VANET, I don't think it's clear that a layer 3 MANET scheme is the future. And there are any number of wireless ad-hoc network concets out there - such as wireless sensor networks (WSNs) often mentioned in IoT discussions. Any type of self-organizing wireless system, not dependent on a base station or other single infrastructure device, would be considered wireless ad-hoc. I'd say that MANET is a subset, because it is specifically an IP-based concept.

I think you have support for this is NetSim. The MAC/PHY are based on 802.22 and you have AODV, DSR, OLSR and ZRP routing protocols

NS2 is no longer used and active development stopped 7 years ago. So its best you use NS3 or NetSim or OPNET. And like you say you need to have lots of time and be very good at programming (in multiple languages) for Ns3.

Yes..Sure

yes infact i am ready with article related to performance analysis of tcp variants in security attacks

 i ll be waiting aitizaz ali!

Hi Afaf

i think this will help you: JiST / SWANS

 

best regards,

ehsan

can you help me that define a wormhole  tunnel between two wormhole nodes in tcl code ?

Hi,

you can go to header file of your protocol for example AODV.h if you are working with AODV. in header file you can find code related to routing table. you have to identify members which stores information of the neighbors. after identifying you simply have to print them in AODV.cc

also go to aodv_rtable.cc find nb_lookup() function it can be used to print neighbour list

You're right Dr Yung. group signatures are perfectly suited for this purposeو but privacy level is not adjustable.

Tracing part is implemented in some protocols that have used ring signatures, such as "Efficient and Spontaneous Privacy-Preserving Protocol for Secure Vehicular Communications" by Hu Xiong et al. It's using trusted party private key. 

Which fading are you talking about? When you say channel model, it seems you are interested in small scale fading. But RSSI has nothing to do with the small scale fading, it is related to large scale fading only.

Your problem statement is not clear. I do not understand what the (x,y) pair is supposed to represent (a node in the network?) and I do not understand what you mean by "fitness" of the node. Fit for what, exactly? What "resource" are we trying to find? What makes one (x,y) location better than another one?

A genetic algorithm is not good for finding a needle in a haystack; it needs some sort of relatively smooth and mostly connected landscape, such that it can start at a random place and then feel its way in a random fashion to take steps uphill to a peak (or in an inverted landscape downhill to a lowest point).

You need to describe your landscape before you start devising your algorithm. Figuring out how to make your landscape suitable for a GA will tell you what your fitness algorithm is: In fact, every GA should begin there, with a landscape that gives you a fitness metric as a single floating point number, and then you can figure out the specifics of the mating and generational dynamics.

for fixed time control you may use Webster algorithm or the method presented in HCM (Highway Capacity Manual). if the traffic flow variation is significant in a time period, actuated or semi-actuated control are useful where the data on flows are obtained by sensors and timing is determined according to these data and some rules, same as martin explained above. new mathematical methods such as fuzzy programming can also be incorporated in the process as well.

Hi,

I think NS2 does not provide an API to create obstacles. You can instead modify the mobility model you're using to "avoid" certain areas in the scenario. Such areas would be the obstacles.

Hope it helps.

Best regards

Hi,

If you install and validate your Network Simulator 2.35, you can use "$HOME/ns-allinone-2.35/ns-2.35/indep-utils/cmu-scen-gen/ns cbrgen.tcl" file with no problem.

You can find detailed usage of cbr traffic generator TCL script as

ns cbrgen.tcl -type cbr -nn 20 -seed 1 -mc 5 -rate 5.0

If you had an error about cbrgen.tcl file permission, you can change your file permission with "chmod 777 cbrgen.tcl" with using your teminal in linux based operating system.

++ I attached a zip file that contains traffic.sh (shell script), and traffic folder (contains cbrgen.tcl and cbr-25-0,cbr-25-10,cbr-25-20). You can use shell script to generate traffic patterns for 0 / 10 / 20 nodes.

I hope that examples will helpful for you. I wish you good luck.

Define the term SELFISH and accordingly set the behaviour of a node. For example, a cooperative node forwards other node's messages and non-cooperative nodes drop other node's messages.

IEEE 802.11 uses the ISM band. More precisely, (possibly overlapping) channels are defined.

The AP and the stations MUST select and use the same channel to transmit and receive their frames. The devices implement CSMA-CA with a half-duplex radio.

Hi Mohammad

See this new published paper on VANET hacking. It discusses various attack scenarios in a practical context. I hope it helps.

You can use following material

Sir I am interested, if still some topics are available

Hi Marwan,

you can find several traces as datasets in my publications list. I have several papers on a mixed indoor-outdoor environment with associated datasets. Feel free to contact me for more information. As an example, see publications attached to this post. Critical can be the fact that we never recorded the elevation/altitude since we are doing 2D localization. However, you can simply add the altitude as 2 m yourself to the traces since the UAV, we used to sense and record data, was keeping the altitude constant at  2m. Good luck.

Hello,

i want to calculate the transmission range of a node in castalia-3.2. But i do not understand how can i do. Pleae can someone help me?

These articles will help you:

Please refer to EDCA in IEEE 802.11e/p.

I think that you need review different types of queue for service of traffic.

First, it depends on the mobility trace (movement pattern) of the nodes in question. Some mobility models more realistically represent some mobility patterns than others. So choosing a realistic mobility model for your scenario is equally important. Second, understanding the difference between on demand and table driven protocols will make a huge difference. So my suggestion is this - study the difference between the two classes of protocols and setup your simulation (preferably with ns-2 or 3) with different scenarios using different protocols and comparing your results.

That is, run the same simulation severally with different protocols (and with even different mobility models if you like) and compare the results. Two of my papers looked into similar area

If it is a blackhole attack and you are doing it on NS2, pl check this link:

If simulation is done on NS3, check this:

I had the same problem. Initially I had the NS2 folder mounted a NTFS directory. Next I moved my NS2 folder to ~/ and it fixed. 

thank you.

I read the document you provided me and I understand that they used the

"collaborative" technique for decoding interleaved Gabidulin codes.

But there is also the conventional Reed-Solomon decoding for

interleaved RS codes, why not use this technique, this is my question.

Is here a difficulty or inconsistency, to decode the Gabidulin codes

not in a "collaborative fashion" ?

Best regards,

It is more easy for you to change them in the underlying files. TCL alone is not enough.

Minimum cell size can be any value, depending on type of application, service, transmit power, number of user, user QoS and bandwidth. The size of cell varies from Macrocell (5 km) to Attocell (1 m) or may be less. The minimum cell size is limited in different network by different parameter. In case of cellular, inter carrier interference (frequency reuse), speed of mobile (eventually number of handover) are the two main parameters, whereas data rate and transmit power in WiFi, and SE (8 bps/Hz), interference in femtocell etc. 

With 4G LTE, some services will run on the 800 MHz spectrum band. This band is close to the frequency used by certain Freeview services. Because of this proximity, viewers who watch Freeview and live close to a new 4G mast, could experience interference. Symptoms for those affected might include a loss of sound, pixelated images or even the loss of all channels. Uplink interference in 4G is another form of interference. Co-channel interference (CCI) and inter symbol interference (ISI) shall be the other forms.  If 4G cell phones move into the newly available lower frequencies at around 600 MHz cable TV will experience interference.

A recent study demonstrated that electrified rail arcing on pantographs could interfere with trackside signalling and control. The emissions were remarkable in the upper UHF to at least 1500 MHz. This suggests that LTE systems in the vicinity of electrified rail may be at risk for uplink interference in the 600 - 900 MHz bands.

4G LTE  Interference may threaten the GPS timing input as well

You can download source code of http://www.wsnlab.org/downloads

Spectrum sensing, spectrum sharing and interference mitigation are one of the defining features of any Cognitive Radio Network. Spectrum sensing is not exhaustive when it comes to interference mitigation in CR. There are many spectrum sensing methods for Cognitive Radio with certain pros and cons, to name a few are energy detection, matched filtering, cyclostationary sensing, wavelet-based sensing, eigenvalue and variance reduction based sensing etc...In principle CR can decide independently whether particular band is free and can be used under specified limitations i.e. sensitivity, tolerable interference etc. However, because of  noise uncertainty in case of energy detection based sensing a CR may not effectively locate the prospective band of interest. Therefore, we resort on cooperation where N cognitive radio cooperatively send their decision to fusion center or base station which based on the decisions of CR and its own decision declares a global decision whether particular band is vacant or not. 

These days almost every standard with CR capability is designed keeping in view the white space database which contains the list of vacant channels based on the information provided by the primary user or secondary user from adjacent CR network. This keeps the CR terminals regulated effectively with negligible interference. 

This may help you.

In a radio environment, simultaneous multipath routing doesn't have sense if the nodes don’t support multiples interfaces in orthogonal frequencies.

You can send a two packets to the same destination but for different paths, but this doesn’t guarantee that a packet doesn’t interfere with the transmission of the other because the medium is shared. If the different paths use orthogonal frequencies, in this case exist a benefit.

I can think of two relatively simple ways to calculate the area of a Voronoi cell.

First, if you know the vertices of the Voronoi cell in cyclic order, you can always use Gauss's Theorem to convert the area (which is integral of 1 over the cell) to a path integral around the cell.  You could also use Green's Theorem to do the same thing, with a different formal result, but the same mathematical result.

Second, you can do the same procedure edge-by-edge in the Voronoi diagram.  Each Voronoi cell can be decomposed into triangles, each associated with a Voronoi edge. Find the length of the Voronoi edge and the length of the (perpendicular) line between the dual nodes, and use this to calculate the area of the triangle associated with this edge; its the same for both Voronoi cells, and you can accumulate these as you loop over the Voronoi edges.

Particle fiter work as follows:

Difference can be in seed, and weght estimation algorithms.

Hi.

There are challenges for an eventual convergence of IoT into the Radio Access Network (RAN) as part of LTE (5G or whatever comes next).

IoT main requirement is energy efficiency, as end nodes are typically energy constrained. While RAN is typically optimised for throughput and mobility. There is no trivial solution for this but several attempts instead. Weightless (from Neul, see: http://goo.gl/Tp24IT) is one road. There are others, e.g. SigFox's ultra-narrow band proposal (see: http://goo.gl/yAVBt8) and LoRa (from Semtech, see: http://goo.gl/OnKmDN).

I'd say main challenges are not in routing. Instead, security and privacy, virtualisation and cloud computing are active research areas related to IoT.

Moreover, the novelty in those existing solutions include: (i) new business models, (ii) operation in unlicensed spectrum, which is free (big savings compared to spectrum leasing) but requires special attention to interference and coexistence in the design of the PHY and DATA LINK layers, and (iii) large radio range and low-bandwidth provides for a relatively much smaller number of base stations to cover a given area, compared to "traditional" LTE RAN.

Hope it helps.

SIB16 (36.331-c41) provides UTC

Hi,  

Read this info as well as the attachment paper. 

my be useful to u. 

IEEE 802.11 RTS/CTS mechanism helps to solve this problem only if the nodes are synchronized and packet sizes and data rates are the same for both the transmitting nodes.

When a node hears an RTS from a neighboring node, but not the corresponding CTS, that node can deduce that it is an exposed node and is permitted to transmit to other neighboring nodes. 

Download source code of OLSR protocol in NS2, all right reserved by Wireless Sensor Networks Laboratory of Iran. Zip and PDF

You can very easily do it using the Satellite toolkit (stk)  sotware. You can simulate LEO constellation and do all types of necessary simulation and analysis on inter satellite link. There is also some linkage possible with STK and OPNET you can find details with AGI.

same answer like Mr Younes. com port used address in 3E8. it is used 9600 bps for speed. if you not synchronize data sampling want to send the data  

I agree with you , but all I  want is how to implement MANET in ns2?

We may use data outliers to detect the abnormal behaviour of a node. 

with 10% load on 100 Mbps link  the throughput would reach upto 10 Mbps and as load or traffic per unit time would be 10,000 packets(frames) per second since each frame of around 1000 bytes.

In fact, DSR can also use PriQueue queue, you need to modify only a few code on it.

DropTaiqueue drops the packets when queue is full. CMU priqueue manages packet at priority manner. 

We did a case study on AODV. Details of the paper as follows:

Journal of Network and Computer Applications

July 2012, Vol.35(4):1249–1259, doi:10.1016/j.jnca.2012.01.019

Intelligent Algorithms for Data-Centric Sensor Networks

Review

Security threats and solutions in MANETs: A case study using AODV and SAODV

Jan von Mulert Ian WelchWinston K.G. Seah

Show more

Abstract

Mobile ad hoc network (MANET) security has become the focus of prolific research efforts. Driven by the unique and considerable difficulties of providing security arising from the dynamic nature of MANETs, many security schemes have been proposed. Rather than trying to encompass the entire field of MANET security, this paper focuses on networks using the popular Ad-hoc On-demand Distance Vector (AODV) protocol and a secure extension to AODV, the Secure AODV (SAODV) protocol. SAODV is representative of a number of secure versions of the AODV protocol in that it relies upon the use of cryptographic mechanisms to protect the routing control messages of AODV from being forged and/or altered by attackers. We conduct a vulnerability analysis of SAODV to identify unresolved threats to the algorithm, such as medium access control layer misbehaviour, resources depletion, blackholes, wormholes, jellyfish and rushing attacks. We then compare this vulnerability analysis to schemes that have been proposed to combat the identified threats. These proposals include multipath routing, incentive schemes, directional antennae, packet leashes, randomized route requests, localized self-healing communities and a reactive intrusion detection node blacklisting scheme.

The "G" flag in an RREQ in AODV has nothing to do with "Granularity", but rather governs certain aspects of "*G*ratuitous" RREPs, i.e., how an intermediate router with a (valid, fresh) path to the destination behaves when receiving a RREQ.

You should look at RFC3561 section 6.3, third paragraph for the "G" flag, and section 6.6.3 for details on "Gratuitous" RREPs. See attached links.

Yes, you can try with Veins + Sumo + Omnetpp.

Digital communications systems can adapt the transmit data rates based on channel state. We call this 'link adaptation' or 'adaptive modulation and coding (AMC)'. For example, if the communication channel is experiencing severe fading, the transmission can be done with robust modulation, e.g., BPSK that offers protection to data in poor channel state in terms of bit error rate; and a channel coding scheme with a low coding rate, i.e., many redundant bits in order to protect the information bits. If the channel state is good, the transmission can be done with 16 QAM and high coding rate that can transmit more bits.

Lookup table is a table that shows the signal to noise ratio (SNRs) ranges in which each modulation and coding scheme (MCS) can be used. There are many types of link adaptation. The lookup table depends on the kind of system you are considering. What is the system?

OPNET supports well know MANET protocols as well as has GUI. However, it requires a license. 

check this tutorial:

What does RED stand for here?

Now go to location

bonnmotion-2.1a/bin/

and type the following line it will display all available model

[root@localhost bin]# ./bm -hm

BonnMotion 2.1a

Available models:

+ Boundless v1.0 Application to construct Boundless mobility scenarios

+ ChainScenario v1.0 Application which links different scenarios

+ Column v1.0 Application to create movement scenarios according to the Column model

+ DisasterArea v1.0 Application to create extended catastrophe scenarios according to the Disaster Area model

+ GaussMarkov v1.0 Application to construct GaussMarkov mobility scenarios

+ ManhattanGrid v1.0 Application to construct ManhattanGrid mobility scenarios

+ MSLAW v1.0 Application to construct MSLAW mobility scenarios

+ Nomadic v1.0 Application to create movement scenarios according to the Nomadic community Mobility model

+ OriginalGaussMarkov v1.0 Application to construct mobility scenarios according to the original Gauss-Markov model

+ ProbRandomWalk v1.0 Application to construct Probabilistic Random Walk mobility scenarios

+ Pursue v1.0 Application to create movement scenarios according to the Pursue Mobility model

+ RandomDirection v1.0 Application to construct Random Direction mobility scenarios

+ RandomStreet v1.0 Application to construct RandomStreet mobility scenarios

+ RandomWalk v1.0 Application to construct RandomWalk mobility scenarios

+ RandomWaypoint v1.0 Application to construct RandomWaypoint mobility scenarios

+ RandomWaypoint3D v1.0 Application to construct RandomWaypoint (3D) mobility scenarios

+ RPGM v1.0 Application to create movement scenarios according to the Reference Point Group Mobility model

+ SLAW v1.0 Application to construct mobility scenarios according to the Self-similar Least Action Walk model

+ SMOOTH v0.1 Application to construct SMOOTH mobility scenarios

+ Static v1.0 Application to construct static scenarios with no movement at all

+ StaticDrift v1.1 Application to construct static scenarios with a drift

+ SteadyStateRandomWaypoint v1.0 Application to construct Steady State Random Waypoint mobility scenarios

+ TIMM v1.0 Application to construct Tactical Indoor Mobility Model mobility scenarios

Now if you want to run particular model (it will display all the available option of specific model )

For Example for RandomWalk

[root@localhost bin]# ./bm -hm RandomWalk

BonnMotion 2.1a

== RandomWalk ========================================================

Version: v1.0

Description: Application to construct RandomWalk mobility scenarios

Affiliation: Toilers - Colorado School of Mines (http://toilers.mines.edu/)

Authors: Chris Walsh

Contacts: bonnmotion@lists.iai.uni-bonn.de

App:

-D print stack trace

Scenario:

-a <attractor parameters (if applicable for model)>

-c [use circular shape (if applicable for model)]

-d <scenario duration>

-i <number of seconds to skip>

-n <number of nodes>

-x <width of simulation area>

-y <height of simulation area>

-R <random seed>

RandomSpeedBase:

-h <max. speed>

-l <min. speed>

-p <max. pause time>

RandomWalk:

-t x (mode: [T]ime. Nodes will walk until time interval x is up)

OR (do not use both -t and -s)

-s x (mode: Di[s]tance. Nodes will walk until distance x has been covered)

I think you should go for Distributed Hash Table(DHT) in MANETs as well as Clustering...... 

Clustering is available as both data mining technique and machine learning technique. Choice depends on, what you want to achieve using clustering?

You can have look at this book : 

Data Mining: Practical Machine Learning. Tools and Techniques, Second Edition. Ian H. Witten and Eibe Frank.

The e-book is available on the web.

Thank you so much Mangesh.

You can use xfuzzy tool instead of using matlab fuzzy logic toolbox. xfuzzy can generate C/C++ or java source code file of fuzzy logic engine designed in its graphical interface. These source code files can be further used in network simulator 2 or Omnet.   

 .

I am working on NetSim can can help you in that

I hope this may help.

Good luck

I have ever seen a real AODV based MANET, which was developed by three chinese students. They integrated a open sourced MANET protocol stack into Android system in a  tablet computer, and powered it by a portable power source. It is amazing, and it is the first real mobile ad hoc network i have ever seen.

Selfish nodes can be said as authorized member nodes of a network who are lazy to co-operate with other node communications. But these nodes are really meant selfish because they will take full benefits of the network services. They accepts packets addressed to it , but drops other multi-hop packets addressed to other nodes. Please go through my publications as it can give you more ideas on this..........................