With the increasing demand for higher connection speeds by home users, higher bandwidth optical networks are making their way closer to the end users. Access networks based on optical components, especially Passive Optical Networks (PONs) provide an efficient mechanism for setting up of these networks. On the other hand, these networks are prone to certain security risks. The study of these risks and solutions for mitigating them is necessary for PONs to become successful. The passive nature of the network also means that some of the security features have to be implemented using novel optical components not requiring O/E or E/O conversions.
This project looks towards identification of security threats in PONs and development of novel optical components for mitigating them.
Current Status
The project started in Fall'06. Currently, we are looking at ways of solving three security problems that have not been completely solved yet at the physical layer, namely:
-
Denial of Service Identification: A large part of mitigating denial of service attacks is to identify the perpetrator. In a TDM PON context, we are looking at users who use another users timeslot for transmitting data, thereby causing denial of service. This topic relates to identification of such users.
-
Denial of Service: After a user is identified as causing denial of service, we want the capability of cutting that user out of the network as soon as possible and restore the network services to other users.
-
Protection: Covers the broad area of loss of service due to fiber damage or breakage. This topic has been covered in current research but is not covered extensively for the case of PONs where the single fiber going from the splitter to each ONU does not provide inherent redundancy necessary for providing protection.