MAGGIE

Visit the website for completed projects.

Projects


Summer 2007  
BGP Black Holes - Dealing with DoS attacks in service provider networks
[Sponsored by Micronet and Nayatel, Pakistan]
[RAs required*]

Traditionally speaking black holes in network are a result of routing errors or incorrect configuration. When a black hole is created in network, any traffic that goes into an interface, does not come out. One can mitigate the effect of DoS attack particularly distributed DoS by creating a black hole to sink all the attack traffic. With a denial-of-service (DoS) attack, in addition to service degradation of the target, there is possible collateral damage such as bandwidth consumption, processor utilization, and potential service loss elsewhere in the network. One method to mitigate the damaging effects of such an attack is to filter the infected traffic and redirect it towards a dynamically created black-hole.

The challenge is to find a way to drop the offending traffic at the network edge, document and track the destination addresses of the redirected traffic, and promptly return these addresses to the service once the threat is averted. Destinationbased IP black hole filtering with remote triggering allows a network-wide destination-based black hole to be propagated by adding a simple static route to the triggering device (trigger). Contact for further details.

Maintaining BGP route consistency and Enhanced Topology Discovery
[Sponsored by Micronet and Nayatel, Pakistan]
[RAs required*]

To identify and visualize inconsistency in routes. Traditional topology discovery tools tend to discover active topology, featuring interconnection between devices. However network administrators also require a bird eye view of the whole network featuring which networks are behind which router. Traditional Topology Discovery Tools return a topology map that is a representation of interconnection of directly connected peers in a network. In a physical topology peers are ports on devices connected to a physical transmission link. At the IP Level peers are hosts or routers, one IP hop away from each other. However, network administrators often want a concise view of the topology that not only shows interconnection between devices, but also a track of which routes/ networks are behind which devices.

The challenge is to develop a software that dynamically discovers the network topology and determines if the advertised BGP routes are consistent with the connected Autonomous Systems.

Automated Network Configuration Implementation
[Sponsored by Micronet and Nayatel, Pakistan]
[RAs required*]

In order to document network configurations and network design, network administrators resort to use tools like Visio and Cisco Configuration Maker etc. The Aim of these tools is not only to document design scenarios and configurations of individual customers, but in addition to it, these tools also aid in understanding of customer specific requirements. In the legacy method; once network is commissioned by one network engineer over his ISP network, he draws a network diagram according to customer requirements. However a mistake in network diagram might lead to incorrect configuration. Since the diagram is being drawn in one tool and configuration is made on other device i.e. no dependency exists between any of these tasks (technically speaking) Also it is worthwhile to note that if a network administrator is provided with the configuration made on devices pertaining to a customer, he conceives, visualizes and draws a network diagram out of it with a some effort.

Thus, the challenge is to develop a tool that can yield network diagram if configuration made on device is provided to it. And if diagram is made using the tool it should yield network configuration to be made on the devices.

Fall 2006  
Traffic Signalling Sytem
[abandoned due to lack of resources]
With Pakistan in context, there is a significant room for improvement in the way the traffic signalling system is managed. Currently the setup is manually controlled. A central observatory is absent. Most importantly the signal timings do not accomodate for a surge in traffic. Personel from the department of traffic police and required to be present in person at a junction if it is overwhelmed with a sudden rise in traffic.



If we could come up with a solution which could observe the flux of the traffic and update the signal timings (independently) for each direction, traffic jams could be avoided. Moreover if the platform is extended to provide a central observatory, we shall not only be able to monitor the traffic flow, but also setup routes for VIPs or emergency vehicles etc.

The concept is in its infancy and we hope to mature it as time progresses.
Digital Fuel Guage
[abandoned due to lack of resources]
With Pakistan in context, the fuel gauges (be it for gasoline or for CNG) in vehicles are analog in nature. Therefore the amount of fuel stored in the fuel-tank is not relfected accurately. Our intention is to devise a mechanism such that the amount of fuel (gasoline or CNG) remaining in the fuel tank could be determined accurately and all changes in the amounts could be measured precisely.
Video Over Edge
[volunteers required]
(Concept by Dr. Tanvir-ul-Haq, CTO Paktel)
With Pakistan in context, we know that mobile phone service providers are enjoying great business opportunities. Another avenue that may be exploited is to develop a service for a video-phone call. These days, all the mobile phones are equipped with a camera and have the support for processing video. Along with this, each hand-set usually has the support for upcoming protocols such as that of EDGE. The intention is to develop a service over EDGE which the users would use to setup a video call.
SPAM
[volunteers required]
Has nothing to do with SPAM mail. A GPS based navigation system for portable devices with value added services.
IP Monitoring for the national backbone
[volunteers required]
(Orchestrated by Dr. Arshad Ali and supported by myself, Dr. Fauzan Mirza and Mr. Ali Sajjad)

Traditional circuit-switched telecommunications is rapidly transforming into the dynamic packet-switching world of IP. One set of applications that has taken immense advantage of this shift and has emerged in the recent years as the popular voice service choice for consumers all around the world is the set of protocols which are used to carry voice signals over the IP network, commonly referred to as Voice over IP or VoIP. Although VoIP suffers the challenges of Delay/Network Latency, Packet loss, Jitter, Echo and Security, its usage is growing exponentially, spurred on by efficient and cheap peer-to-peer solutions by the likes of Skype and Gizmo, etc.

As the popularity of VoIP is growing and PSTN users are switching to the cheaper VoIP in increasing numbers, governments are becoming more and more interested in regulating ad monitoring VoIP in a manner similar to legacy PSTN services. So, to analyze IP network traffic for billing purposes, and to prevent revenue leakage in large telecommunication organizations, a high performance IP-traffic monitoring system is the need of the hour, which has the capability to provide packet capturing/analysis at all the OSI Layers as well as the ability to identify various traffic usage patterns of importance at the core network level, without adding significant delay in the communication.

A Platform Solution addresses network wide needs, such as security, billing and lawful intercept, etc. The hardware and software costs can be shared among several foundation applications, thus reducing cumulative costs. Up-front planning is required to maximize the effectiveness of data collection for all applications serviced, thus reducing operational and support costs. It is obvious that a Platform Solution is essential. Various international vendors are providing this kind of solution, which provide software used by telecommunications services providers to optimize their IP platforms and protect networks against malicious attacks, analyze network traffic flow, and monitor network activity. However, the disadvantages of using an outside vendor’s solution for meeting PTA’s requirement are manifold.

We aim to provide a Platform Solution called “LAST VOICE” that shall satisfy all possible requirements of our intended clients. The solution will consist of specialized off-the-shelf network hardware equipment for traffic and flow capturing and storage in addition to an indigenously-built software traffic scrutiny engine which will provide comprehensive and real-time in-depth analysis of the monitored network protocols, applications and services. In addition to this analysis, it will carry the functionality of blocking the traffic flow of choice e.g., VoIP or P2P traffic and also deduce the source and destination of such traffic with miscellaneous meta information. Furthermore, a comprehensive Reporting functionality will be provided that can be used for generating Revenue Analysis and Reports in addition to managing QoS provision and usage guarantees to and from various subscribers of the services provider.
Human Resource Management System
[volunteers required]
(As defined by wikipedia) "Human Resource Management Systems (HRMS, EHRMS), Human Resource Information Systems (HRIS), HR Technology or also called HR modules, shape an intersection in between human resource management (HRM) and information technology. It merges HRM as a discipline and in particular its basic HR activities and processes with the information technology field, whereas the planning and programming of data processing systems evolved into standardised routines and packages of enterprise resource planning (ERP) software. On the whole, these ERP systems have their origin on software that integrates information from different applications into one universal database. The linkage of its financial and human resource modules through one database is the most important distinction to the individually and proprietary developed predecessors, which makes this software application both rigid and flexible".

The intention is to develop a solution initially for NIIT which can later be customized for other organizations.
Summer 2006  
RF Motion Detector
[abandoned due to lack of resources]
Current speed traps use laser range finders to determine the speed of a moving vehicle. They have an inherent flaw that the device must be aimed at the oberved vehicle. Our intention is to devise a speed trap that employs the doppler effect using the reflections obsserved by a CW radar. By observing the red and blue shifts (of the reflections) we intend to determine the speed, acceleration, direction of motion and the size of the vehicle. We are using off-the-shelf components to develop the solution.

Last updated: 20th January, 2008.