Venu Madhav Kuthadi

Ant-based distributed denial of service detection technique using roaming virtual honeypots

Nowadays, distributed denial of service (DDoS) becomes a major challenge in the network as it affects the network at multi-level. This leads to traffic overhead and wastage of bandwidth utilisation. In order to overcome these issues, ant-based DDoS detection technique using roaming virtual honeypots is proposed. In this technique, virtual roaming honeypot along with the multi-level secure architecture is used to collect the information about the various intruders at different levels in the network. Ant colony optimisation technique is used to detect the intruders based on the pheromone deposit on that considered area. A multi-level IP log table is used to detect the intruders at different levels of the network. Once the affected area is found, the information is sent to multi-level architecture to limit the spread of the affected area to the honeypot. This information is sent to the honeypot to make a defence system against the attackers. The advantage of the proposed technique is that it provides a full defence against DDoS at multi-level without creating any traffic overhead.

An Enhanced Security Pattern for Wireless Sensor Network

Numerous schemes for data aggregation are done by encryption for privacy of data, and thus, homomorphism has been reviewed and designed on the wireless sensor network (WSN) to improve security. WSN is an emerging and challenging technology in the area of the research because of its vital scope with low power energy associated with it. Several application sensors collect information or data from different sensed nodes and will be aggregated to a host computer or a base station. Data aggregation happens in a network between intermediate nodes which causes to reduce energy consumption on these nodes to make efficient network performance. The algorithm used in existing aggregation system does not provide sufficient security functionality and is more vulnerable to several attacks. Furthermore, some compromised nodes also inject false data which leads to a falsify data aggregation which is forwarded to base station. The enlargement of WSNs is thrombosed due the limited energy constraints. The main focus of this paper is done mainly on enhancing energy in WSN by working on the enhancement of some routing protocol. This paper also explains on eliminating of data vulnerability, security aggregation, and false data injecting attack by presenting different robust surveys and protocols. In this paper, initial discussions are done on WSN with a detailed overview of sensor, and reviews are made on providing security of wireless sensor network. The proposed novel idea is used in aggregating data during exchanging of messages and to preserve its data privacy and also to overcome problem in network construction (NC) and its security. The NC is done by clustering its topology, and assigning heads to each cluster with huge communication range, also security pattern generation is created for protecting the data information.

Honey Pot: A Major Technique for Intrusion Detection

Generally, Intrusion detection system (IDS) is installed in industrial environment for protecting network that works based on signature, where they are not capable of detecting most unidentified attacks. The detection of undefined attack and intrusion is not more helpful to identify the several kinds of attack, where intrusion-based attack has become a challenging task to detect intruder on network. A skilled attacker can obtain a sensible information and data from the system after knowing the weakness. Distributed denial of service (DDoS) is a major thread over the security and most enlarging thread in recent days. There are so many types of Denial of Service (DoS) such as Teardrop, Smurf, Ping of Death, and Clone attack. The aim of the cyber defense system is to detect the main cause of the several counter attacks on the enterprise network. On the way to fix these issues, we are proposing a novel idea that relies on honey pot technique and packet data analysis which are trained by the sample of malware after using the Intrusion detection technique in both ways separately as Network and Anomaly intrusion detection system. Some approaches are not being easily implemented in the network of real enterprises, because of practicability training system which is trained by the sample of malware or deep analysis of packet inspection or depends on the host-based technique that requires a big capacity for storage over the enterprise. The honey pots are one of the most successful techniques to collect the sample of malware for the purpose of analysis and identification of attacks. Honey pot is a novel technology which consists of massive energy and possibilities in the field of security. It helps reading the behavior of the attack and attacker information.

An efficient adaptive preprocessing mechanism for streaming sensor data

Wireless Sensor Network (WSN) is fixed in many sensing environments to capture and monitor events. The sensing values come from sensor device that may contain noise, missing values, and redundant features. Noise, missing values and redundant features should be removed from the streamed data using an efficient preprocessing mechanism and then preprocessed data can be provided for further processing such as classification or clustering. If any errors occur in the streaming data then the preprocessing mechanisms should be able to handle the errors adaptively. Many preprocessing techniques are implemented for preprocessing streaming data, to adapt dynamic changes, and to handle different situations. The problem is the preprocessing systems does not efficiently handles different situations and adapt to changes for streamed sensor data. In this research, a new adaptive preprocessing mechanism is proposed that will efficiently handle changes in the incoming streaming data and scenarios are implemented to decouple the preprocessor and predictor in different situations for increasing the prediction accuracy. The proposed system uses PCA (Principal Component Analysis) as preprocessor and Hyperbolic Hopfield Neural Network (HHNN) as predictor. This method provides an efficient and adaptive preprocessing of streaming data.

Energy Efficient Secure Data Transmission in Wireless Sensor Network

Wireless sensor network (WSN) turns out to be a vital innovation identified with systems as well as it is by and large broadly connected and received in the present day control and checking functions. In any case, problems with respect to vitality proficiency create huge worries during the time spent in remote sensor system. For problems concerning vitality, power efficient adaptive clustering hierarchy (PEACH) tradition has been suggested essentially for WSNs so as to criticize vitality use of hubs and expanding life range of the system. The hubs in the remote sensor system structure as a gathering. The most noteworthy vitality hub in the gathering has been chosen as a bunch head. The bunch head determination has a few criteria in choosing: (1) based on the vitality levels of sensors in group (i.e., the hub with higher vitality would be chosen as a bunch head). (2) based on hub degree (total number of hubs associated with it) (i.e., the hub with advanced hub degree would be chosen as a bunch head). The time differential of arrival (TDA) has been proposed for vitality consumption of sensor hubs in the system. TDA decreases the vitality utilization of hubs and expanding the system life duration. The node connectivity algorithm (NCA) has been proposed to discover the availability with neighbor’s hub in the remote sensor system. The bunch head confirms the gathering hubs and transfers the confirmation data to server. On the off chance that any noxious hub available in the system, the server would recognize through self-mending key appropriation procedure and the bunch head would shut the hub and toss out the hub from the bunch. The noxious hub assets would be stored into the server. It will make the system more ensured and safe one. To broadcast the parcels from resource to goal, figure the most limited way with the assistance of Dijkstra algorithm. The bundle encodes at source and unscrambled once it gets to the goal by means of Advanced Encryption Standard (AES) encryption calculation.

Hybrid Fuzzy Recommendation System for Enhanced E-learning

The heterogeneous e-learning materials are generated in the progress of online e-learning technique. The system of e-learning is providing huge opportunities for learning online for learners with enhanced and efficient practices of learning. The system of e-learning needs to cater for the individual learner requirements including learner’s profile and activities of learning in the form of tree structure. There are several issues of pedagogical learning. In case of learning phenomenon, this is too difficult for any learner or user to select their suitable learning resources without having exact background knowledge. To address these issues, this research is proposing two enhanced techniques called as Hybrid Fuzzy-based Matching Recommendation Algorithm and Collaborative Sequential Map Filtering Algorithm. This proposed approach recommends a new method to assist users on their individual as well as collaborative learning methods for accessing learning resources.

Stealthy Attacks in MANET to Detect and Counter Measure by Ant Colony Optimization

MANET (Mobile Ad Hoc Network) is the active wireless network which can produce communication less connections in which every node can perform as a router. Every node in the MANET is responsible for dynamically creating other nodes for communication. Protection is an important examine for the wireless and wired network communications in MANET. The achievement of MANET (mobile ad hoc networks) strappingly depends upon people’s assurance in its sanctuary. Though, uniqueness of MANET poses both opportunity and challenges in the accomplishment of security purpose, like authentication, privacy, access, availability, integrity, non-repudiation, and control. The malicious node presence is being a major challenge in MANET for designing the vigorous security solution which can defend the MANET from different routing attacks. We have categorized the existing security services and security threats in this mobile ad hoc network paper. The proposed approach gives a better solution in the achievement of present security threats in MANET. Based on monitoring system a guard node is recommended in this paper through which every activity occurred inside the network can be constantly monitored. If there is any abnormal action of Stealthy attack occurred inside the network then the Guard node collects and forwards the abnormal information toward the source node. In this guard node an algorithm based on Ant Colony optimization is deployed to recognize the threats which are present in the MANET’s network. This proposed system detects the stealthy attack and also resists the attackers from the network. Therefore, this approach makes the enhancement in the network packet delivery and its performance ratio, etc.

Data Size Reduction and Maximization of the Network Lifetime over Wireless Sensor Network

The main concept of this research is for increasing the network lifetime and decreases the data size over wireless sensor network. To perform this idea we proposed some novel technique which provides the reliable energy efficient routes and maximizing the network lifetime for finding the route that minimize the total energy for packet traversal. We also use the data compression model that reduce the size of data and joint balancing of nodes and optimize the dynamic compression for improving the lifetime of network. The data compression could be completed within some step, those are raw data could get broken in few branches and get compressed at distinct level of the compression, these compressed data could be decompressed at a certain level and again compressed with distinct level to forward directly or by using some base station. For transmitting the data to base station from the source node, every node has to be clustered and have to choose one cluster head in the group of every cluster, the CH (Cluster Head) is having the more energy in compared to the all other nodes. The CH (Cluster Head) is obtaining the entire message from the other neighbor’s nodes and transmits it to the Base station. From source to destination data transmission, the nodes are searching shortest path that provide a high computation of complexity.

EIDPS: An Efficient Approach to Protect the Network and Intrusion Prevention

Nowadays, Network Security is growing rapidly because no user specifically wants his/her computer system to be intruded by a malicious user or an attacker. The growing usage of cloud computing provides a different type of services, which leads users to face security issues. There are different types of security issues such as hacking intrusions worms and viruses, DoS etc. Since the entire needed resources are associated with everyone and are centrally monitored by main controller in cloud computing area it creates a simplest way for intruders. In addition, an experienced or knowledgeable attacker can get to make out the system’s weakness and can hack the sensible information or any resource and so, it is essential to provide protection against attack or intrusion. Additionally, to handle poor performance or low latency for the clients, filtering malicious accesses becomes the main concern of an administrator. Some of the traditional Intrusion Detection and Prevention Systems fail to overcome the abovementioned problems. As a result, this research proposes a novel approach named Enhanced Intrusion Detection Prevention System to prevent, protect and respond the various network intrusions in the internet. Our proposed system use client-server architecture, which contains main server and several clients. Clients record the doubtful actions taking place in the Internet and record the suspicious information. Then, this recorded information is forwarded to the main server for further usage. After that, the main server analyses the received data and then make decision whether to provide a security alert or not; then the received information is displayed via an interface. In addition, server verifies the data packets using IDPS and classifies the attacks using Support Vector Machine. Finally, as per the attack type Honeypot system sends irrelevant data to the attacker. The experimentally deployed proposed system results are shown in our framework which validates the authorized users and prevents the intrusions effectively rather than other frameworks or tools.

Hybrid Technique for Frequent Pattern Extraction from Sequential Database

Data mining has became a familiar tool for mining stored value from the large scale databases that are known as Sequential Database. These databases has large number of itemsets that can arrive frequently and sequentially, it can also predict the users behaviors. The evaluation of user behavior is done by using Sequential pattern mining where the frequent patterns extracted with several limitations. Even the previous sequential pattern techniques used some limitations to extract those frequent patterns but these techniques does not generated the more reliable patterns .Thus, it is very complex to the decision makers for evaluation of user behavior. In this paper, to solve this problem a technique called hybrid pattern is used which has both time based limitation and space limitation and it is used to extract more feasible pattern from sequential database. Initially, the space limitation is applied to break the sequential database using the maximum and minimum threshold values. To this end, the time based limitation is applied to extract more feasible patterns where a bury-time arrival rate is computed to extract the reliable patterns.