Taner Tuncer

Indoor localization with Bluetooth technology using Artificial Neural Networks

The most important function of a sensor network is to collect information from the environment. For many applications, it is important that the location or sensor that originates the collected information is ascertained. This article presents the detection of a mobile sensors location in an indoor environment with the help of known location sensors (anchors) placed in the environment. Anchor sensors measure temperature, which is sent to a mobile phone via Bluetooth. The mobile phone can measure RSSI values of incoming signals as well as the temperature information coming from each of the anchor sensors. The Artificial Neural Network(ANN) model presented in this article was developed to detect the mobile phone location. The ANN model accepts the Received Signal Strength Indicator (RSSI) measured by the mobile phone and the anchor sensor ID number as inputs. The ANN was first trained and tested, after which the error between mobile phone locations obtained in test results and actual locations was calculated. The results were compared through the Centroid Localization (CL) method, as is known in the literature. According to the results thus obtained, it was shown that more accurate location detection was possible with the ANN model.

The implementation of ASG and SG Random Number Generators

Linear Feedback Shift Register (LFSR) is mostly used in the implementation of Pseudo Random Number Generator (PRNG). LFSR based on PRNG techniques are used for many applications such as generating data, encryption keys and generating padding bits. Although a majority of random number generators have been implemented in software level, hardware implementation is becoming more and more popular due to the advent of faster and high density Field Programmable Gate Arrays (FPGA). In this paper, we propose implementations of FPGA Shrinking Generator (SG) and Alternating Step Generator (ASG) using LFSR based random number generation. Both systems have been implemented on Altera Cyclone IV board, and random number has been generated in the real time. Generated numbers have been tested according to National Institute of Standards and Technology (NIST) statistical test. According to the results, both SG and ASG have been shown to able to use in the cryptographic systems.

Detection DoS Attack on FPGA Using Fuzzy Association Rules

This paper proposes a programmable embedded system based on data mining and fuzzy logic to determine Denial of Service (DoS) attacks in real time. The proposed system detects the DoS attacks to a web service. The system consists of two phases. At the first phase, the number of terminated connection and the number of connection request statistics and so on have been extracted from packet. These obtained features constitute training data. Each record in training data was obtained with 2 sec intervals from network traffic. Then Fuzzy classification rules have been obtained from training data by data mining and fuzzy logic. At the second phase, DoS attacks have been detected using these rules in real time for testing purpose. The proposed system has been tested on Alter a Cyclone III EPC3C40F484C7 in the FPGA environment.

FPGA based programmable embedded intrusion detection system

In this article, the design of a Field Programmable Gate Array (FPGA) based real time Intrusion Detection System (IDS) is presented. The embedded system first summarizes and normalizes the packets and then determines whether there is an attack or not using this summary according to the naive bayesian algorithm. The packet summaries which play an important role in the accurate classification of the packets include data such as the IP numbers of the packets, port numbers, protocol types and packet length. Real time Intrusion Detection System (IDS) was designed in the Altera Cyclone III based FPGA development environment and tested. As a result of the tests performed, appropriate results were obtained.

Hybrid PRNG based on Logistic Map

Very well know discrete time chaotic random number generator used in cryptographic applications is Logistic Map. In this paper, we present the implementation of hybrid random number generator based on Logistic Map on Field Programmable Gate Arrays (FPGA). The numbers generated by the hybrid system have been obtained at 32 bit floating point notation. For the usage of cryptographic applications of obtained numbers, these values have been normalized to 0 and 1. Total 20000 numbers have been generated by hybrid system, and they have been tested according to NIST(National Institute of Standarts and Technology) test suite. According to the test result of the numbers obtained from hybrid system, it has been shown that can be used in cryptography applications.

Programmable embedded system design:A study on sniffer

Intrusion Detection Systems (IDS) are designed for capturing the attacks from internet or local area network consisting of various packets and data. Software or hardware IDSs detect the attack considering the network and user parameters. In order to determine the attacks, it is necassary to know some properties of the packets which convey data between networks. Capturing the packets between Networks and computers is called as sniffing. The MAC address, packet size, packet type and IP addresses of the captured packets are given to IDS an input. Thus, the first step is to capture the packet for designing the IDSs. It is very crucial to perform these procedures in real time and fastly. The implementation of real time IDS with a programmable embedded system is a very attactive topic. In this paper, the procedure of capturing the necassary packet and frame is presented to implement a real time IDS programmable embedded system.

Random number generation with LFSR based stream cipher algorithms

Random numbers have a wide range of usage area such as simulation, games of chance, sampling and computer science (cryptography, game programming, data transmission). In order to use random numbers in computer science, they must have three basic requirements. First, the numbers generated must be unpredictable. Second, the numbers generated should have good statistical properties. Finally, the generated number streams must not be reproduced. Random number generators (RNGs) have been developed to obtain random numbers with these properties. These random number generators are classified into true random number generators (TRNG) and pseudo random number generators (PRNG). One of the PRNGs used for generate random numbers is Stream Encryption algorithms. In this paper, random number generation of LFSR based stream encryption algorithms and their hardware implementations are presented. LFSR based stream encryption algorithms have been implemented on Alteras FPGA based 60-nm EP4CE115F29C7 development boards by using VHDL language. The obtained random numbers passed the NIST statistical tests, accepted as standard for cryptographic applications.

Frequent pattern mining for community dedection in web logs group based habit dedection in community using network traces

During last 10 years, internet usage network has spread at an unforeseen rate. Concurrently, many services including official transactions have been granted from the internet. By the way, it has become an area where internet users have access to apps and interact with each other over social media. Peoples internet usage habits have become a domain that can give information about the areas of the population and their interests. In social networks, group analysis and connection predictions are popular terms that construct base of a lot of scientific works. This terms, especially, used for; ads customize, habits and tendency retain or determination of friendship based on same character. Network users can be grouped according to their activity and character, or they can be grouped according to their movement over the network by their estimation algorithms. In this study, all dataset filtered to get anonymous traffic logs that generated by a part of internet users who are in the network of an institution and the community was explored. All logs analyzed according to Category-based diversity, time periods of accesses and usage periods. As expected, communities were established from the habits of users with “Pattern Based Frequency Analysis Method”. Voluntary experimental users whose mac addresses were already defined and belong to determined groups redistributed with specified method. Lastly, via calculating the achievement of correct distribution results the success of method found out.

Determination of location using RSSI and LQI based on fuzzy logic

Sensor location estimation is an important problem in wireless sensor networks. GPS can be used to obtain location information. GPS systems are not ideal for many applications due to their excessive power consumption and high cost. As an alternative to GPS, Sensor location can be estimated through the usage of at least 3 nodes with known locations. Received Signal Strength Indication (RSSI) is the simplest and most inexpensive technique used to determine distance and location, and is a standard feature on every sensor. But, only the use of RSSI is not accurate. In addition to the RSSI parameter, LQI is used too. However, RSSI can be affected by noise and environmental obstacles Kablosuz ağın konumlandırıldığı ortam homojen bir yapıya sahip olmadığından. For this reason, it is difficult to set up a mathematical model for RSSI. this paper present a method based on fuzzy logic to estimate the location of a sensor of unknown location. Estimate location simulation was made in Matlab and obtained results were compared with centroid localization method.