Computer Networks

Hi, I can suggest you the following papers: 1. Analytical model of failure in LTE networks http://ieeexplore.ieee.org/xpl/articleDetails.jsp?reload=true&arnumber=5431443 2. Analytical Modeling of the LTE Radio Scheduler http://link.springer.com/chapter/10.1007%2F978-3-658-00808-6_7 3. An analytical model to study the impact of time-varying cell capacity in LTE networks http://ieeexplore.ieee.org/xpl/articleDetails.jsp?tp=&arnumber=6097234 4. A Semi-Analytical Macroscopic MAC Layer Model for LTE Uplink http://ieeexplore.ieee.org/xpl/articleDetails.jsp?tp=&arnumber=6208726 5. Analytical Analysis of the Coverage of a MBSFN OFDMA Network http://ieeexplore.ieee.org/xpl/articleDetails.jsp?tp=&arnumber=4698234 Anyway, I agree with the previous comment. So many components must be considered, and for this reason I think it is very difficult to find an analytical model for LTE networks which covers all the aspects (e.g., radio failure, channel variations, channel dependent scheduling, user mobility, QoS profiles, time and frequency domains scheduling).

Hi, if it's HCCA that you need, you may want to take a look at this: http://cng1.iet.unipi.it/wiki/index.php/Ns2hcca#Documentation it's the 802.11e HCCA module for ns2.

While striving after a definitive (whatever it means) answer to the question what kind of science computer science is, I think we also have to consider what Abelson, H. and Sussman, G. J. argued about in their book entitled Structure and Interpretation of Computer Programs (MIT Press, 1996, second edition). According to them, 20th century computer science was striving for a “procedural epistemology” (while computing was trying to use and exploit procedure as knowledge). It seems that now, at the beginning of the 21st century, we have arrived to a junction. I think in the coming (not too many) decades, computer science will be engaged with seeking an "information epistemology" as well as a "system epistemology". These may operationalize computing in a completely different way: computing will be an integral (synergistic) part of holistic systems. Whichever way it goes ahead, this new “computer science” will get closer to studying phenomena that are associated with nature, but also with social formations and engineered socio-technical systems. "Computing" (again whatever it means) will be ‘just’ part of systems, without its own ‘teleology’. As systems are becoming parts of the fabric of the natural, social, technical and/or cognitive worlds, informing is becoming much more important than information, and information is becoming the means of having order, rather than that of communication (as described by Shannon). Information that resides in physical, biological, social, human, and engineered systems will be addressed in a comprehensive and holistic way, and that will drastically influence “computing” in systems, such as next generation cyber-physical systems. Putting all these together we can forecast that the already observable trend of (possible and probable) systems evolution will take us back to the investigation of natural physical, biological, social and human phenomena. Having these, and many more, thoughts in my mind I do not think it is wise to take any firm position. It will be temporary. I am very much sure that 21st century "computer science" will subsume "procedure" as a foundational concept, but it will also extend it with the also fundamental concepts of "information" (informing) and "system" (adaptation/evolution), considering their dialectic relationship with both natural (physical, biological, social and human) and created environments. Does it mean that 21st century computing will be an epistemology of non-linear systems with cyber physicality? Perhaps. Probably. We do not know.

There is effectively no difference between pseudo-parallelism and real parallelism from the possible concurrent accesses point of view. The only difference is that things will go faster with real parallelism. Anyway, in some rare cases with pseudo parallelism, we can have potential concurrent accesses (write accesses for example) to a given resource from several threads, but this will luckily work! This is because in pseudo-parallelism (on a single CPU), even if we have concurrent accesses, they are sequential. The problem is when a task or thread commutation occurs when a thread did not achieve to terminate its write access before its execution quantum ends. So if you are lucky, a parallel program could work without generating coherency problems. But this is pure luck! When things really work in parallel (real parallelism on multiple CPUs or on a multi-core CPU), you have to be very very lucky for your program to work correctly without any synchronization techniques. This is because without synchronization, there is not any kind of sequential execution among threads. To be honest, in every case, you need to use synchronization techniques like mutual exclusion to ensure that you program will correctly work. Being lucky, is never a good choice for parallel programming :-)

If you need to generate graphs automatically then GraphViz is very good: http://www.graphviz.org/

I will followup with Osama's question. "Simulated vs Analytical Results". What about any one of this compared with "Real-world Results". The reason for this question is that, sometimes for the real-word cases you do not have strong mathematical deductions, however, one is able combine Simulation studies and then proceed to implement a real-world testbed. Does it mean that that result might not be as strong compared with Analytical results?

Not really a simulator, instead a library for developing your own simulations: ns2. But only for fault tolerance.

Wireless networks are being used a lot nowadays and more and more new means are being developed to attack it and even if there are not many attack incidents, due to rise of use in various power-grid and mobile applications, etc, it is expected to invite more attackers. And as someone mentioned above, it is about disrupting a service, wireless or wired. In wireless, DDoS works more like a Jamming attack and is quite easier to launch. You might wanna look into wireless attacks in more detail and evaluate its similarity with other attacks which are simpler to launch.

Thanks v.r.sarma Dhulipala ,Mustafa Mustafa andTeacher Tien Do! Need to a lot time to understand :(

WiMAX standard is nt providing any meschanism for the scheduling but its the responsiblity of the operator how to manage the resource. as the in mobile WiMAX the channel status is not predicitable so we can apply adaptive modulation.

un texto que puede ser de interés: http://es.pdfsb.com/readonline/5a564245656731305633523144586c6d56454d3d-2896656

You can access the RSSI from the physical layer parameters.

Thank you ,I hope you help me always.

Yes the hand-off. You are right about two wireless cards, but if you use virtual machine configured (using bridged networking) on physical system then that behaves like a physical system. You can add more wireless cards and you play with more then one wireless card.

Thank you Sirapat

Capturing and analysing data flow on the own comouter with protocol analyzer, like Wireshark (freemail tó download) is good exercise in practice.

VLAN and LAN are two terms used frequently in the networking field. “LAN” is abbreviated as “Local Area Network” is a computer network to which a large number of computers and other peripheral devices are connected within a geographical area. VLAN is an implementation of a private subset of a LAN in which the computers interact with each other as if they are connected to the same broadcast domain irrespective of their physical locations

Thank you Amitabh and Steve Steve: graphs are sparse. a "dens subgraph" is a subgraph like or near a clique but it is not a clique and we can make a clique from it by adding some edges.

Here is a simple comparison: - Filtering based approach is about filtering traffic(malicious) by using firewalls or other filtering mechanisms. - Capability based security is about binding security to a token/key. Any user who holds this key will have the capability to access particular system resources/services. - Overlay based security is about hardening the network by using overlay networks a.k.a. virtual network. I think SDN (Software Defined Networking) can be applied in such cases.

NS-3 provides you with the most up-to-date interface to simulating network protocols. Opnet is a commercial alternative.

You may use jpcap with java but in java text processing in tough so you may try python with any network data capturing tool