With the increase of digital data on the internet, computers are at higher risk of getting corrupted through cyber-attacks. Criminals are adopting more and more sophisticated techniques to steal sensitive information from the web. The botnet is one of the most aggressive threats as it combines lots of advanced malicious techniques. Detection of the botnet is one of the most serious concerns and prominent research area among the researchers. This paper proposes a detection model using the clustering algorithm to group bot traffic and normal traffic into two different clusters. Our contribution focused on applying K-means clustering algorithm to detect botnets based on their detection rate (true and false positives). Experimental results clearly demonstrate the fact that with the help of clustering we were able to separate the complete dataset into two entirely distinguishable clusters, where one cluster is representing the botnet traffic and other one representing the normal traffic.

Intrusion Detection is one of the most common approaches used in detecting malicious activities in any network by analyzing its traffic. Machine Learning (ML) algorithms help to study the high dimensional network traffic and identify abnormal flow in traffic with high accuracy. It is crucial to integrate machine learning algorithms with dimensionality reduction to decrease the underlying complexity of processing of huge datasets and detect intrusions within real-time. This paper evaluates 10 most popular ML algorithms on NSL-KDD dataset. Thereafter, the ranking of these algorithms is done to identify best performing ML algorithm on the basis of their performance on several parameters such as specificity, sensitivity, accuracy etc. After analyzing the top 4 algorithms, it becomes evident that they consume a lot of time while model building. Therefore, feature selection is applied to detect intrusions in as little time as possible without compromising accuracy. Experimental results clearly demonstrate that which algorithm works best with/without feature selection/reduction technique in terms of achieving high accuracy while minimizing the time taken in building the model.

Less contiguous nature of military applications demands for surveillance of widespread areas that are indeed harder to monitor. Unlike traditional Wireless Sensor Networks (WSNs), a military based large size sensor network possesses unique requirements/challenges in terms of self-configuration, coverage, connectivity and energy dissipation. Taking this aspect into consideration, this paper proposes a novel, efficient and secure clustering method for military based applications. In any clustering based approach, one of the prime concerns is appropriate selection of Cluster Heads and formation of balanced clusters. This paper proposes and analyzes two schemes, Average Energy based Clustering (AEC) and Threshold Energy based Clustering (TEC). In AEC, a node is elected as Cluster Head (CH) if its residual energy is above the average energy of its cluster whereas in case of TEC, a node is elected as Cluster Head if its residual energy is above the threshold energy. Further, both AEC and TEC choose nodes as CHs if their distance lies within safety zone of the Base Station. In this paper, aim is to come up with a solution that not only conserves energy but balance load while electing safe nodes as CHs. The performance of proposed protocols was critically evaluated in terms of network lifetime, average residual energy of nodes and uniformity in energy dissipation of nodes. Results clearly demonstrated that AEC is successful in incorporating security whilst increasing overall lifetime of network, load balance and uniform energy dissipation.

Energy efficiency is a very crucial issue for battery operated Wireless Sensor Networks (WSNs). Routing plays a major in energy dissipation and it is shown in the literature that Cluster based approach is the most energy effective in any network in comparison to direct or multi hop based approach. Therefore, optimized Clustering became a key point to achieve energy efficiency in Wireless Sensor networks. In this paper, we have designed and implemented a novel protocol in MATLAB in which Cluster Heads are chosen on the basis of energy threshold, minimum average distance from surrounding nodes and farthest distance among Cluster Heads to provide optimal coverage. This paper also compare results of randomly selected CHs and farthest CHs and results demonstrates that farthest chosen CHs provide much better results than randomly selected CHs. To further evaluate performance of our protocol, results of our protocol are compared with LEACH and proposed protocol dominates LEACH in terms of minimizing transmission distance, energy dissipation and hence increasing network lifetime. Apart from this, proposed protocol is based on Poisson distribution because simulation results clearly states that Poisson distribution is very well suited for WSN in comparison to Uniform and Random distribution.

Wireless Body Area Network (WBAN) is one of the most critical real time networks. In order to increase network lifetime, several protocols have been proposed in the Literature. Majority of these protocols are based on either coverage or residual energy of the network. However, none of the protocols is based on node criticality, communication count, coverage and residual energy. In this paper, we have taken into consideration all the above stated parameters to determine the optimal path for transmission of message. In particular, we are implementing this scenario for Athlete Activity Monitoring to identify the athlete with maximum physical stamina. As per best of our knowledge, only intra-BAN routing protocols are proposed till date. However, in this paper we have simulated Athlete Activity Monitoring scenario for intra-WBAN and inter-WBAN architecture. It is clear from the simulation results that our scheme performs much better in terms of energy consumption than earlier routing schemes.

Wireless body area networks (WBANs) are emerging as one of the newest forms of Wireless Sensor Networks. In WBANs, sensor nodes accumulate human physiological data and transmit it to the sink node. However, transmission of physiological data to the sink node over a mobile route becomes a very daunting task for sensors due to their limited battery power. Moreover, replacement of critical sensor nodes is a major challenge in such scenarios. In order to increase network lifetime, some routing protocols have been proposed in the literature, but the majority of them are focused on coverage distance and residual energy of sensor nodes. In this work, we will propose an energy efficient routing algorithm for WBANs that will take into consideration two additional attributes (node criticality and communication count) along with the above stated attributes. These four parameters in the proposed algorithm will result in enhancement of working lifetime of the network via less energy consumption in comparison to existing routing protocols.

Underwater Wireless Sensor Networks (UWSNs) explore aquatic environment to facilitate various underwater surveillance applications. However, UWSN unique features also impose new challenges such as limited bandwidth, huge propagation delay, mobile nature of nodes and high error rates. UWSNs deployment in unattended environment further exacerbates their vulnerabilities to the attacks. These challenges make security solutions proposed for Wireless Sensor Network (WSN) impractical to be applicable for UWSN. This paper analyzes the problem of security and mobility in UWSN and proposes Cluster based Key management Protocol (CKP), a new key management protocol for hierarchical networks where sensor nodes form cluster around more capable nodes. CKP also proposes a new communication architecture that handles mobility efficiently and minimizes the impact of a node compromisation to itself. CKP provides confidentiality, authentication, integrity and freshness. The performance evaluation demonstrates that the CKP is energy and storage-efficient. Further, we investigate the survivability and the security of the CKP against various security threats in order to evaluate its effectiveness.

Investigations in hydrologic sciences are bounded because most of existing water surveillance methods are manual. Such systems are incapable to gather information at spatial and temporal level due to location constraints. In this paper, we present a new communication architecture for Underwater Wireless Sensor Network (UWSN) based on acoustic communication. However, unique challenges offered by underwater environment are main hindrance in deployment of UWSN in real life applications. So, we discuss various communication methodologies to figure out which one suits best to the requirements of UWSN. Our simulation results illustrate change in data transmission rate, energy consumption and transmission time w.r.t transmission range and number of nodes in network. Results prove that multi-hop communication offer high transmission rate, large bandwidth. Moreover, multi-hop communication is much more energy and time efficient than direct communication.

Increased employment of WSN (Wireless Sensor Network) in real life applications and their hostile and remote locations accelerate demand of security in WSN. Publicly accessible wireless communication channel also makes WSN vulnerable to numerous security attacks. Scarcity of resources acquaints new sort of challenges and difficulties during implementation of effective security mechanisms. In this paper, we evaluate and compare performance of three different security mechanisms (ECRKS, CKP and AP scheme). ECRKS (Energy-efficient, Connected, Resilient Key pre-distribution Scheme) is based upon multi hop communication architecture specifically designed for homogeneous WSN. Clustering based protocols, AP (Asymmetric pre-distribution) scheme and CKP (Clustering based Key management Protocol) are proposed for heterogeneous WSN. All the above mentioned schemes are simulated in MATLAB to evaluate their effectiveness and suitability for WSN. Simulation result shows that CKP outperforms other two schemes in terms of transmission distance, memory burden, energy dissipation and resilience.