No abstract

Emergency Call Alert (ECA) Application is developed for inform the urgency as well as seriousness of the call to the receiver. The profile in the mobile facilitates that the user to change the mode they need and this requires manual work load. This Call Manager changes the pro-file automatically to the desired mode designed by the user. It has also been enhanced with filtering mechanism called as Emergency calls filter which allows the calls from the emergency list to be in user preferred mode irrespective of the current mode. This feature is to avoid the missing of calls from the numbers specified in the emergency call list. Now a day most of calls come to us when we are busy, it is an interrupt, which disturbs or sometimes accident may happen. To avoid this either we switch off the mobile or we put the mobile in silent mode. If this is so if any emergency call comes we cannot attend such type of calls. The other disadvantage of this is the calling person does not know when the called person is free. To overcome all these problems we presented paper, Emergency Call Manager which handles all the call when he/she is busy sends an SMS. Software which filters this message and it gives an emergency alarm to called mobile.

A mechanism that is intended to expose information against a security violation in a network is the use of network covert channel and it is difficult to detect information about data loss like location of loss using network covert channel. To identify the covert channel were the data pattern missing over the sharing of resources in networks. Several mechanisms are used to identify a large variation of covert channels. However, those mechanisms have more limitation like speed of detection, detection accuracy etc. In this paper, a new machine learning approaches called “Support Vector Machine and Hyperbolic Hopfield Neural Network” to overcome the drawbacks of existing methods. This approach is supported to classifying the different covert channels with data packets which is shared in networks and its supports to identifying the location of data loss or data pattern mismatched. Finally, the proposed methods properly detected covert channels with high accuracy and less detection high speed shared a network resources in effective manner.