Computer Communications

This data set is for running simulation for NDN and mostly on my CCndnS caching policy. some data for random and popularity based caching policies are provided as well. The name of the folders indicates the cache size of each Content Store (CS). Depends on topology of the network (Abilene, Chained and Chained with branches), there are different set of sources of files and content requesters. The measured values from client side is saved in a file with name ClinentX or Client_Rx. The data related to each router is presented in a file with the name of the router. For Abilene network, you can find the topology and the name of the routers in my paper. Be noted that in my paper routers' name starts from 1 (R1) but data files router name starts from 0. So R1 in the paper is R0 in the data set. There is one more file in each folder which provides the general information about the simulation like the cache size of each router, total Interests generated in the simulation or total network hit probability which is the probability of finding data from the network regardless of which router cached data. The name of some of the folders is like "alpha0.8". That means the parameter alpha for zipf distribution is set to 0.8 in this set of simulations. Folders with name "SLA" belongs to Service Level Agreement application that I explained in my paper. Many parameters can be found with value 0. These are parameters that I deactivated in this set of simulations but in general they can be measured. I used them for other studies.

data from three entire days: (1) before Black Friday (Nov/21/2014); during Black Friday (Nov/28/2014); and (3) after Black Friday (Dec/05/2014). There are different sections of network traffic: (1) normal; (2) DDoS; (3) Flash Crowd; and (4) Flash Crowd + DDoS which is a very rare scenario. Since the IP of hosts are confidential, they are masked by letters (one letter represents just one number). The data in this repository is related to site P (78 samples with around 500,00 records each). The dataset is available in the CSV format (text). The fields are limited by apostrophe, separated by semicolon and compressed with 7z. Mirror at: https://uspbr-my.sharepoint.com/:f:/g/personal/and1970_usp_br/EvRKMh0d3slPpDkdZgBuR-oBULsPbgnHvnsqC9P5UzBFOQ?e=TAgPyc . A complete version of this dataset with all the e-commerce sites can be found at: https://uspbr-my.sharepoint.com/:f:/g/personal/and1970_usp_br/EvYstyiZiWdEhOfiI21IOWwB1lCHXsrNUKPpBx7-z1rnWQ?e=efkCVa Please in case of use, cite our original researches: 1) Grouping Detection and Forecasting Security Controls Using Unrestricted Cooperative Bargains; 2) A Proposal to Distinguish DDoS Traffic in Flash Crowd Environments.

The four datasets include the evaluation of various opportunistic routing protocols in four datasets (Reality, Dartmouth, Lyon, INFOCOM2005) using the Adyton simulator (https://github.com/npapanik/Adyton). More specifically, the datasets include the performance data (performance metrics), the processed data used to build the decision matrix and the scoring results for the examined routing protocols using different combinations of decision-making and weighting methods. The four datasets correspond to the four experiments presented in the manuscript entitled "A framework for the evaluation of routing protocols in opportunistic networks". The file ranking-tables.xlsx contains an example highlighting the pitfalls in using traditional performance metrics when ranking protocols using MCDM methods. The example ranks the performance of six well-known opportunistic protocols in the Dartmouth dataset using the SAW decision-making method and the MW weighting method. There are two rankings, one produced using traditional performance metrics and the other using the proposed normalized metrics. The example corresponds to tables 2,3,4 and 5 of the manuscript "A framework for the evaluation of routing protocols in opportunistic networks".

Spectrum scarcity is a critical problem that may reduce the effectiveness of wireless technologies and services. To address this problem, different spectrum management techniques have been proposed such as overlay cognitive radio (CR) where the unlicensed users can share the same spectrum with the licensed users. The main challenges in overlay CR networks are the identification and detection of the Primary User (PU) signals in a multi-source narrow-band interference (NBI) scenario. Therefore, in this paper, we investigate the performance of an orthogonal frequency division multiplexing (OFDM) overlay CR network with Secondary Users (SUs) and subcarriers selection schemes. Three approaches for SUs and subcarriers Selection named Direct, Distributed and Incremental selection techniques are proposed in this paper to increase the expected signal to interference and noise ratio based on full or partial knowledge of the channel state information (CSI). We also show that Distributed selection techniques provide all the SUs equal chances to be selected without affecting the selection diversity gain. General as well as simplified outage probability expressions are derived and extensive simulations are conducted to evaluate the performance of the proposed techniques and support the theoretical derivations. To accommodate more SUs, a new approach for asynchronous NBI estimation and mitigation in CR networks is investigated. Without any prior knowledge of the NBI characteristics and based on sparse signal recovery theory, the proposed approach allows the PU to exploit the sparsity of the SUs interference to recover it and approach the interference-free limit over practical ranges of NBI power levels.