International Journal of Communication Technology for Social Networking Services Vol. 1, No. 1 (2013), pp.13-16 http://dx.doi.org/10.21742/ijctsns.2013.1.1.02 Soft Computing Approach to evaluate Trust Value of a Node in VNET with HMM through Supervised Learning Jitendra Singh Sengar BVM College of Technology & Management, Gwalior [email protected] Abstract Soft computing approach to find the trust value of the node over the VNET, This paper present trust value evaluation of the sent message over the network to check their whether the sender is valid or verified. Through this paper we got trust value in percent by which it is easy to decide what appropriate action should be taken. Here we have a trained model through which we could analyze the node authentication and their trust value. In that model user may enrolled as authentic node or can get trust value by showing their ID proof and authentic document to get proper identification. Keywords: TVEP Supervised Training Model, VNET, ANN, HMM 1. Introduction Whole idea occurs from Hidden Markov Modal, Node’s Trust Value evaluation In VNET through supervised learning approach of artificial neural network. Three combinational efforts are used to identify the node in VNET by which node could judge about it authentication in VNET. When a new node enter in VNET node will capture Through there attribute then proposed algorithm is applied to got attribute to identify the node. Node will have to register before getting entry in any ADHOC network zone to get UID from current ADHOC Zone Through which node could travel safely. Proposed algorithm will Automatically assign UID & alter id to just assigning their status ID value when a node move from one ADHOC VNET to another the node will get the UID on the behalf of information or behavior got from the node behavior in previous ADHOC VNET than our proposed algorithm will active their working on the behalf of assigned unique id pattern, that pattern will completely identify and recognize through our combinational effort of these three approach. 1.1 Hidden Mrkov Modal (HMM) Sequence of emissions observe by HMM, but do not know the states sequence, the model generate the emissions. It got update from observations which come in the form of ratings after direct experiences or recommendations requested from intermediaries. 1.2 Trust Value Definition 1.2.1. Direct Trust: Established from the previous interactions with the trustee. 1.2.2 Recommended Trust: Measureable from quantifying the trustworthiness of the trustee in a given context and at a given time point as communicated by intermediary. 1.2.3. Reputation Trust: Trust Value could get through various ways from attribute or combinational attribute of VNET attribute and Node attribute those are currently being used that concept is important for communication and network protocol designers when ISSN: 2205-863x IJCTSNS Copyright ⓒ 2013 GV School Publication International Journal of Communication Technology for Social Networking Services Vol. 1, No. 1 (2013) creating trust relationships between participating nodes. VNET ADHOC system allots specific reputation degree of ID to ambulance or police or army vehicle. 2. Moving Pattern Control 2.1 Auto Assign Value As we divide the network into regions consisting of grids, an optimal moving pattern for the inter- and intra region levels must be selected. For the intra region level, we select an extension of the town hall method. Each grid elects a commonly trusted moving representative, and these nodes move to the capital to exchange intra region trust information. 2.2 Supervised Learning Approach ANN The learning approach of artificial neural network is efficiently execute in the supervised manner i.e., the whole way through which entire result is to be occur or come out from the possible input provided i.e., the result & the way to train a network. A algorithm is proposed in complete supervision of trained structure. 3. Proposed Structure & Algorithm In this proposed algorithm, node identification & node behavior could be identify on the behalf of their trust value which is evaluated through combinational approach of HMM and Supervised Learning , the whole pattern is drawn to identify its entire attribute. That string pattern will work as UID of node in a particular VNET in which node is currently available. The trained system will check the entire words of allotted specified value from VNET, which is emitted through hidden markov model, and then the training is done on the behalf of each word meaning and after getting the proper value at proper place. It would help to evaluate the percent of trust value correctness if the trust value evaluation got 100% result then assigned trust value to this node is one i.e. the node is reputed and it will eligible to get trust certificate from VNET. The entire process of proposed methodology is shown in above figure SFigure 1.1, Proposed pattern: SBMP07SA6184 Figure 1. Supervised Learning Applied on Specified Assigned Pattern 14 Copyright ⓒ 2013 GV School Publication International Journal of Communication Technology for Social Networking Services Vol. 1, No. 1 (2013) 3.1. Proposed Algorithm to Execute Entire Strategy 1. Get input string entered node in VNET. 2. That input string will be on the Network manager guideline (Specified uid number assigned from VNET). 3. Emit each word from specified assigned pattern [i.e., SBMP07SA6184]. 4. After 3rd step perform apply whole patterned string to supervised trained model to get their trust value. 5. If TV=100% then the node is recommended or authentic. 6. End Each associated substring has their importance to make the complete specified pattern of VNET in which node is available. Because allotted pattern from VNET has entire basic attribute of node through which VNET can easily maintain secure network among node. Through this system no unwanted node can affect the system. That will reduce the chance of accident or unauthorized node’s activity. If the node is in VNET it will got UID from VNET manager our proposed pattern to VNET manager is like SBMP07SA6184 that character or numeric values combination build from attribute of the node through which node can be authorized to travel safely in VNET. Trained System will Evaluate Trust Value in Percent as Shown in Table 1. Table 1. Pattern Execution in Trained Modal to Evaluate Trust Value in Percent S. No. 1 2 Substring of pattern S SB TV in % 0 10 3 4 5 SBM SBMP SBMP0 10 25 25 6 7 SBMP07 SBMP07S 50 50 8 9 SBMP07SA SBMP07SA6 75 75 10 11 SBMP07SA61 SBMP07SA618 75 75 12 SBMP07SA6184 100 SBMP07SA6 SBMP07S SBMP0 SBM S 120 100 80 60 40 20 0 SBMP07SA6… TV in % TV in % 1 2 3 4 5 6 7 8 9 101112 SG. - Analytic Graphical Representation of Entire Approach Copyright ⓒ 2013 GV School Publication 15 International Journal of Communication Technology for Social Networking Services Vol. 1, No. 1 (2013) 4. Conclusion The work provide a more secure and better way to maintain safe & secure journey of moving node in VNET, Here we have a supervised intelligent algorithm which trained through emission of each word step by step from hidden markov modal. After emission of specified and assigned patterned trust value which is identified and verified through trained and supervised neural network for their authentication after identifying proper authentication node could travel in VNET otherwise node will not get permission to enter in network. Future work the specified pattern would be more complex and would be more secure from unauthorized access. Approach of this proposed combinational pattern work would help to establish more secure and safe journey while travelling from one place to another. References [1] [2] [3] [4] [5] [6] [7] 16 R. Ismail Josang and C. Boyd, “A Survey of Trust and Reputation Systems for Online Service Provision”, Decision Support Systems, vol. 43, no. 2, (2007), pp. 618-644. W. Zhang, S. Das and Y. Liu, “A Trust Based Framework for Secure Data Aggregation in Wireless Sensor Networks”, Proceedings of Annual IEEE Communications Society Sensor and Ad Hoc Communications and Networks, vol. 1, pp. 60-69, (2006). S. Buchegger and J. Boudec, “Performance Analysis of the Confidant Protocol”, Proceedings of International Symposium on Mobile Ad Hoc Networking and Computing, pp. 226-236, (2002). P. Michiardi and R. Molva, “CORE: A Collaborative Reputation Mechanism to Enforce Node Cooperation in Mobile Ad Hoc Networks”, Proceedings of IFIP TC6/TC11 Sixth Joint Working Conference on Communications and Multimedia Security, pp. 107-121, (2002). M. Grossglauser and D. Tse, “Mobility Increases the Capacity of Ad-Hoc Wireless Networks”, Proceedings of IEEE INFOCOM, (2001). B. Liu, P. Brass, O. Dousse, P. Nain and D. Towsley, “Mobility Improves Coverage of Sensor Networks”, Proceedings of International Symposium on Mobile Ad Hoc Networking and Computing, (2005). S.I. Singh and S.K. Sinha, “A New Trust Model using Hidden Markov Model Based Mixture of Experts”, International Conference on Computer Information Systems and Industrial Management Applications (CISIM), pp. 102-107, (2010). Copyright ⓒ 2013 GV School Publication
© Copyright 2026 Paperzz