Indrajit Bhattacharya

DirMove: direction of movement based routing in DTN architecture for post-disaster scenario

Network architecture based on opportunistic Delay Tolerant Network (DTN) is best applicable for post-disaster scenarios, where the controlling point of relief work is any fixed point like a local school building or a hospital, whose location is known to everyone. In this work, 4-tier network architecture for post-disaster relief and situation analysis is proposed. The disaster struck area has been divided into clusters known as Shelter Points (SP). The architecture consists of mobile Relief Workers (RW) at tier 1, Throw boxes (TB) at tier 2 placed at fixed locations within SPs. Data Mules (DM) like vehicles, boats, etc. operate at tier 3 that provide inter-SP connectivity. Master Control Station (MCS) is placed at tier 4. The RWs are provided with smart-phones that act as mobile nodes. The mobile nodes collect information from the disaster incident area and send that information to the TB of its SP, using DTN as the communication technology. The messages are then forwarded to the MCS via the DMs. Based on this architecture, a novel DTN routing protocol is proposed. The routing strategy works by tracking recent direction of movement of mobile nodes by measuring their consecutive distances from the destination at two different instants. If any node moves away from the destination, then it is very unlikely to carry its messages towards the destination. For a node, the fittest node among all its neighbours is selected as the next hop. The fittest node is selected using parameters like past history of successful delivery and delivery latency, current direction of movement and node’s recent proximity to the destination. Issues related to routing such as fitness of a node for message delivery, buffer management, packet drop and node energy have been considered. The routing protocol has been implemented in the Opportunistic Networks Environment (ONE) simulator with customized mobility models. It is compared with existing standard DTN routing protocols for efficiency. It is found to reduce message delivery latency and improve message delivery ratio by incurring a small overhead .

A Co-operative Approach to Thwart Selfish and Black-Hole Attacks in DTN for Post Disaster Scenario

A Delay Tolerant Network (DTN) is prone to security threats, especially the Selfish and Black-hole attacks. In this work, we have taken up these two security issues in a DTN based on an architecture suitable for Post disaster scenario, where mobile nodes move around a fixed destination and communicate messages to it. A co-operative scheme is presented to reduce the harmful effects of these malicious nodes in the network. The cooperation is obtained from the neighbouring nodes. The malicious behavior of a node is examined at the time of sending messages to that node by inquiring other neighbouring nodes about the past performances of that node. Simple parameters, such as total number of sent and received messages to and from other nodes and the elapsed time since the network is active, are used for arriving at the decision of forwarding messages to that node. The node whose malicious behavior is being tested is not involved in the whole process as it may manipulate information to pose as a valid node. By the cooperation of other neighbouring nodes, a combined faith value (CFV) is calculated. If this value satisfies a pre-defined threshold condition, then the node is treated as valid for that moment. The process is performed every time a node wants to transmit a message to another node. Also, the calculated CFV is multicast in the neighbourhood for updating. The scheme has been tested on the Spray and Wait (S&W) DTN routing protocol with mobile nodes and fixed destination type architecture to prove its effectiveness. It is found to avoid malicious nodes and increase the trustiness of the network and data communication.

On design and implementation of a scalable and reliable Sync system for delay tolerant challenged networks

Seamless data synchronization among peer entities is a problem for any challenged network scenarios where infrastructural supports for synchronization are not available or insufficient. Such a network follows a delay/disruption tolerant approach, in which connection or contact time among devices are intermittent and short-lived, agents may use devices of heterogeneous nature with different communication and processing powers, and the communication channel may have high loss rate that impacts the application layer file synchronization protocols. The bundle protocol specifications for delay tolerant networks only describe the semantics for data transmission between two distant devices, and do not specify how a peer-to-peer (P2P) file synchronization protocol can cater seamless data exchange among multiple heterogeneous communication devices. We propose in this paper a new P2P sync, called pSync, on the top of the bundle protocol which precisely takes care of prioritized file Sync with role based transfer applicable for challenged networks. Our testbed experiments conducted with information dropbox, ground and aerial data mule suggest that pSync is scalable and clearly outperforms BitTorrent Sync in delivering even 30% more priority packets in certain test cases.

Embedding an Extra Layer of Data Compression Scheme for Efficient Management of Big-Data

Use of Smartphone as mobile nodes in different communication infrastructures is excessively explored in recent years. Such smart phones can be considered as a good candidate for situations like Disaster Management, where there is no infrastructure available to support communication and connectivity among the group members is a prime objective. Disaster rescue operations are generally based on location intensive operations including neighboring nodes’ locations and their availability. The storage limitations of such devices ask for suitable strategies to store information efficiently. In this work, a method has been proposed that employs an extra layer of compression, while storing location data in the form of latitude-longitude (lat-long) pairs, to the HBase database. Location data in a mobile network is big-data, as continuous collection of such information adds numerous data inputs. By incurring a negligible overhead on the system in the form of small encoding and decoding time, the proposed method obtains almost 70 % compression ratio, even for thousands of input data. In this work Huffman lossless encoding scheme has been used.

Parameter Selection and Performance Study in Particle Swarm Optimization

The present paper describes the Particle Swarm Optimization (PSO) technique and how different parameters in the algorithm may be selected in order to achieve faster convergence to the solution for a given optimization problem. PSO has become a common heuristic technique in the optimization community with many researchers exploring the concepts, issues and applications of the algorithm. PSO has undergone many changes since its introduction in 1995. As researchers have learnt about the technique, they have derived new versions, new applications and published theoretical studies of the effects of the various parameters and aspects of the algorithm. This paper comprises a snapshot of the particle swarming, including variations in the algorithm, current and ongoing research, and applications. In this paper we first analyze the impact that the inertia weight and maximum velocity have on the performance of the particle swarm optimizer, and then provide guidelines for selecting these two parameters.

Analysis and Early Detection of Rumors in a Post Disaster Scenario

The use of online social media for post-disaster situation analysis has recently become popular. However, utilizing information posted on social media has some potential hazards, one of which is rumor. For instance, on Twitter, thousands of verified and non-verified users post tweets to convey information, and not all information posted on Twitter is genuine. Some of them contain fraudulent and unverified information about different facts/incidents - such information are termed as rumors. Identification of such rumor tweets at early stage in the aftermath of a disaster is the main focus of the current work. To this end, a probabilistic model is adopted by combining prominent features of rumor propagation. Each feature has been coded individually in order to extract tweets that have at least one rumor propagation feature. In addition, content-based analysis has been performed to ensure the contribution of the extracted tweets in terms of probability of being a rumor. The proposed model has been tested over a large set of tweets posted during the 2015 Chennai Floods. The proposed model and other four popular baseline rumor detection techniques have been compared with human annotated real rumor data, to check the efficiency of the models in terms of (i) detection of belief rumors and (ii) accuracy at early stage. It has been observed that around 70% of the total endorsed belief rumors have been detected by proposed model, which is superior to other techniques. Finally, in terms of accuracy, the proposed technique also achieved 0.9904 for the considered disaster scenario, which is better than the other methods.

Mitigating selfish, blackhole and wormhole attacks in DTN in a secure, cooperative way

In this work, security threats posed by selfish, blackhole and wormhole attacks in a delay tolerant network (DTN) are addressed. A cooperative approach has been proposed as a solution where malicious behaviour of a node is tested whenever another node tries to send a message to it. Performance parameters, like received and sent messages to neighbouring nodes have been used along with calculated faith values (CFV) to derive latest CFVs. The process is repeated for each communication instance. In addition, a simple yet efficient credibility-based cryptographic encryption technique is applied to encrypt message contents that works by appending the signatures and CFV values of each encountered nodes. The method has been tested, by implementing it over the spray and wait DTN protocol, in the ONE simulator. The proposed strategy is able to avoid the three types of malicious nodes effectively and increases the trustworthiness of communication process greatly.

An Efficient and Secured Routing Protocol for VANET

Since few years, Vehicular Adhoc Networks(VANET) deserve much attention. Routing and security are the two most important concerns in this type of network. A number of routing protocols already exist for vehicular ad hoc network but none of them are made to handle routing and security issues side by side. In this paper we propose a new junction-based geographical routing protocol which is capable of dealing with routing as well as security issues. This protocol consists of two modules: (i) To implement routing, at first this protocol selects appropriate junctions dynamically, through which a packet must transmit in order to reach to its destination. (ii) To deal with security issues, we have introduced the concept of mix-zones to prevent vehicle tracking by unauthorised users. At the end, the performance of the proposed work has been compared with some well known existing routing protocols depending upon some parameters like packet delivery ratio and normalized routing load in a simulated environment.

Comparative Performance Analysis of DTN Routing Protocols in Multiple Post-disaster Situations

A Delay Tolerant Network, DTN in short, presents a practical solution as providing an effective response and monitoring of a post-disaster situation. A lot of DTN routing protocols exist that have workable applications in a post-disaster environment. The current work presents a comparative performance analysis of various routing protocols in the field of DTN like the Epidemic, the Prophet, Spray and Wait, MaxProp and DirMove over three different post-disaster mobility patterns that have been modeled for performance analysis in different environments. Important performance parameters like delivery ratio, delivery latency, and overhead have been measured for deriving the comparative results. Current work compares and then proposes a DTN routing strategy that can be a generalized solution for implementation in any type of disaster situation. Simulation results show that DirMove routing protocol is best suited and most efficient to perform in a post-disaster type situation having predictable mobility patterns.

CTMR-collaborative time-stamp based multicast routing for delay tolerant networks in post disaster scenario

Due to high chances of loss in connectivity, a Delay Tolerant Network (DTN) can be used to communicate between nodes without having any fixed connection between the source and destination. A Post Disaster Scenario presents a very challenging environment to communicate in and to analyze those situations is even a harder task to accomplish. It necessitates very efficient co-ordination to successfully accomplish situation analysis and resource management. Efficient co-ordination between relief and rescue teams in such situations can be achieved through multicasting, since it allows sending single packet to multiple destinations. Though multicasting in MANET has been studied extensively, but the implementation of efficient multicasting in DTN is a very challenging task due to its frequent partitioning characteristic. In this work a Collaborative Time-Stamp based Multicast Routing (CTMR) Protocol has been proposed. The messages have been implemented using customized bundles, where the destination of a multicast bundle consists of a group of nodes. Node grouping mechanism has been adopted to suit a post disaster condition. Collaborative bundle creation and selection mechanism has been utilized so that localized redundant information flow is minimized. Suitable time and space limits have been selected to further reduce redundancy. The group forwarding strategy is based on probabilistic measures calculated using historical encounter records, and on a multiple parameter priority queue. This protocol has been implemented in the ONE simulator and is compared with other existing unicast and multicast routing protocols on important routing parameters like delivery ratio and delivery delay. Comparison results show that CTMR can be a novel efficient solution to multicasting in a DTN.