Click on a topic to view the introduction to the project
- 1. Vector Sensor Receivers in Underwater Networks details
Mentor: Prof. Ali Abdi
Over the past few decades, only the scalar component of the acoustic field, i.e., the pressure, has been used for communication in underwater channels and networks. In this project we take advantage of the vector components of the field, such as the three components of acoustic particle velocity, in addition to the scalar pressure, at the receiver. With small size vector sensors that can be efficiently manufactured today, one can receive the data over these unexplored degrees of freedom of the channel. The main advantage of such a system is its low error probability, without using a large-size pressure-only receive array. This is particularly important in systems which have serious size limitations.
Using a Matlab-based underwater channel simulator, the student will simulate different types of underwater channels. Then some channel modeling issues in velocity channels of vector sensors will be investigated. Furthermore, several different receiver structures using vector sensors will be simulated in underwater channels. Participating students should have skills in Matlab programming and the basics of propagation and communication in underwater wireless networks.
- 2. Micron-Nano Screens and Their Applications details
Mentor: Prof. Haim Grebel
A detection problem is encountered whenever narrow spectral lines signature, characteristic of a specific chemical agent, is distributed over a broad spectral range: it necessitates use of a scanning spectrometer and data are collected for a long period of time. Instead, we have used a set of pre-determined ‘finger-print’ filters in the infrared (IR), Terahertz (THz) and sub-millimeter spectral regions. This concept is general and may be applied to screens in the nano-scale.
Micron size screens which are imbedded with nano-size, self-assembled array of spheres (Fig. 1) exhibit rich spectral characteristics for bio-detection, remote sensing and imaging applications. For the most part, simulations as well as fabrication are made in-house at NJIT using a variety of simulation and fabrication tools.
Fig. 1. Micron imbedded self-assembled array of nano-size spheres for special purpose spectroscopy
The REU students will be interfacing with a group of doctoral and post-doctoral student as part of their study. They will be involved in computer simulations, sample characterizations and other exciting activities in the laboratory: for example, these screens have been recently incorporated with carbon nanotubes – a novel crystalline structure at the nano-scale.
- 3. Intrusion Detection Systems details
Mentor: Prof. Edwin Hou
An intrusion detection system (IDS) is one of the front line defense systems that a network administrator can deploy to guard against network attacks. However, the volume of information generated by an IDS is typically enormous, making it difficult to analyze. Furthermore, the attack scenarios can be hidden within large set of logs and messages which hinder the identification of the attack. One solution to this problem is to automate the search for clues in the logs and build a semantic network that describes the various attack scenarios. This provides a quick way to identify network attacks and analyze the attack scenario. This project will investigate the use of Petri-nets techniques for building and analyzing the semantic network.
The undergraduate research component will involve compiling attack scenarios from attack logs, extracting principal components of the attack scenarios to form semantic networks and use Petri-nets techniques to detect network attacks. The student must have a strong logical background and programming capabilities.
- 4. Design and Implementation of iSpace via Networking, Sensing and Control details
Mentor: Prof. Mengchu Zhou
Intelligent Space (iSpace) is a relatively new concept. It intends to use distributed sensors, actuators, computing processors, networks, the related information technologies, and other intelligent entities like robots to enhance the space such that it offers higher productivity environments for the human being and other resources. In factories, it means improved product quality and production rate. In offices, it means improved efficiency for people working in such space. This project intends to define 1) what one can gain by using wireless and sensor networks in iSpace design and development; and 2) what additional technologies, e.g., sensors, actuators, human-machine interfacing technologies, and intelligent machines, need to be developed for the purpose. We are especially interested in using 1) available image acquisition devices (e.g., webcams) and processing capability to ensure that the space is in a good shape; and 2) available human-machine interfacing technologies to facilitate office workers in obtaining the needed help via Internet and getting informed of any information related and only related to the roles they are playing. Students are required to be creative and energetic with knowledge in computer programming and to have good hands-on capabilities. Algorithms for data and knowledge mining and complex system control need to developed and demonstrated.
- 5. Characterizing the Impacts of Emerging Distributed Denial-of-service Attacks on Streaming Traffics details
Mentor: Prof. Nirwan Ansari
Internet is increasingly playing a significant role in modern economies, and hence its security is vital. Attacks on the Internet are becoming more sophisticated and difficult to detect. Recent denial of service (DoS) attacks such as low rate DoS attacks and Reduction of Quality (RoQ) attacks are demonstrating this trend of increased sophistication. At the Advanced Networking Laboratory, we are focusing on characterizing effects of these attacks on different types of Internet traffic. We showed that these attacks have adverse impacts on the VoIP traffic quality in the Internet. We are now investigating their effects on the streaming media traffic in the Internet. We typically use ns2 simulator to characterize the attack effects.
For the NSF Undergraduate research program, the REU candidate will work with graduate students in conducting simulations of different attack models and then studying various adverse scenarios. The REU candidate will need basic networking protocol knowledge and C/C++ programming skills to participate in this project. The candidate will be given early exposure to the field of Internet security that would foster his/her future academic or industry research career.
- 6. Design and Evaluation of Wavelet Transforms details
Mentor: Prof. Ali Akansu
Wavelet transform has been forwarded in the literature over two decades ago and successfully applied in many engineering fields including multimedia, biomedical and communications. This project will deal with designing orthogonal function sets and their evaluations with respect to time and frequency domain performance metrics relevant to the application at hand.
- 7. Hardware Implementation for a Sensor Networks based on Chemical Sensors details
Mentor: Prof. Durga Misra
Chemical sensors based on surface acoustic wave (SAW) demonstrate excellent characteristics to be implemented in wireless sensor networks. In this project students will design an interactive signal conditioning circuit along with a communication strategy for the built in network controllers and transceivers that analyzes and monitors chemical concentration profile remotely in a system-on-a-chip platform. The signal conditioning circuits will be implemented to make the SAW sensor, integrated into a self-sustaining system. Adaptive digital filtering will be used as a deconvolution procedure in the sensor that will have advantages over cross-correlation such as signal to noise ratio, faster identification of absorbent and ability to analyze changing samples. The immediate interface circuit with amplifiers along with the SAW device will be implemented using phase shift keying approach. This circuit takes into account the loss of 20 to 40dB introduced by the SAW sensor into the signal. The attenuation of the SAW sensor depends on the Q-factor of the interdigital transducer and due to the conversion of electric waves to acoustic waves by piezoelectric effect in addition to surface losses that occur on the surface of the substrate and chemical coating. The interface circuit, therefore, detects the frequency shift (0.5-1 KHz) as a function of the chemical species concentration absorbed by the sensor and alerts the network controller when a monitored parameter exceeds some threshold, based on local processing and measurements. Circuits will be designed in a mixed-signal environment. HSPICE simulation of the circuit as a function of frequency shift will be determined and the circuit can be optimized for different chemicals and concentrations.
- 8. Enabling Next Generation Connectivity through Ubiquitous Social Computing details
Mentor: Prof. Quentin Jones
Ubiquitous computing technologies (smart phones, WiFi, RFID tags, sensor networks, navigation systems embedded in cars, etc.) and social software systems (flickr photo sharing, instant messaging, Wikipedia, etc.) are changing people’s everyday collaboration and coordination in ways, and on a scale never seen before in human history. An obvious yet poorly understood way in which ubiquitous social computing applications could improve our social environment is through artfully designed and deployed social recommendation applications. These recommendation applications could be in the form mobile social match making systems that recommend new friends/colleagues to users in physical proximity, or location-aware social-venue recommendation systems that help people find good places to go and engage with others. However, such applications bring potential risks as well as benefits. Technology can be overwhelming and counterproductive when people must constantly master a multitude of heterogeneous gadgets, each running different software. And people can be concerned and rights actually violated, if social issues such as trust and privacy protection are not carefully and artfully designed into ubiquitous social computing applications.
REU students will have the opportunity to join one of several research groups creating and researching ubiquitous social recommendation systems. Students will work in the Smartcampus lab were researchers are building a location-aware community system test-bed designed to serve as a dispersed laboratory for the study of people-to-people-to-places services in terms of social connectivity, co-ordination and privacy. The test-bed is enabled through the provisioning to 500+ students with wireless, locatable devices (primarily PDA cell phones), running a suite of mobile social software.
- 9. Studying Fragility of Routers and Servers details
Mentor: Prof. Roberto Rojas-Cessa
Today’s Internet infrastructure involves the use of routers and appliances in the network side, and servers and clients in the user side. Both sides of the network are prompt to failures and malicious attacks. However, it is not well understood how sensitive network infrastructure actually is.
As part of this research project, the student will perform laboratory experiments on a network testbed to study the fragility of network routers and servers. The experiments require the student to become familiar with network equipment, including programming of routers, servers, and high-throughput traffic generators. This project involves a great amount of laboratory and experimental work. The student will work with graduate students and participate in technical meetings to discuss work and findings.
The successful student is required to have a solid background on networking, TCP/IP, operating systems (Windows and Linux), probability, and statistics.
