Recent Research Areas
MIMO and Cooperative Communications and Networking
Ahmed K. Sadek, Wipawee (Pam) Siriwongpairat, Ahmed Ibrahim, Karim Seddik, Zhu Han, Andres Kwasinski, Thanongsak (Kee) Himsoon, Weifeng Su, Wei Yu, Amr El Sherif, Quoc Lai
It is amazing to witness in recent years the huge proliferation of wireless applications with high demands in terms of signal quality, data rates, and coverage requirements. Many emerging applications require a collection of infrastructure-less nodes to communicate with each other self-configuring themselves into a wireless network as in Mesh networks, sensor networks, or more generally mobile ad hoc networks (MANETs).
The scarcity of conventional resources as energy and bandwidth has driven the efforts of the researchers to develop new innovative communications and signal processing schemes that can face this unprecedented growth of wireless applications. One of the new concepts that emerged recently and gained wide attention is that of cooperative communications. Cooperation is a new communication paradigm that completely changed the way researchers view wireless networks. In this paradigm, users share there antennas to forward each others information, hence forming a virtual antenna array. This new form of distributed spatial diversity offers more reliable communications, increased network capacity, extended coverage area, more energy efficient communications, etc..., therefore becoming a very good candidate to face the growing demands in wireless applications.
This new exciting research area poses a lot of challenging questions to researchers. In our research group, we have diverse interests in addressing many of these important and fundamental questions. We list here some related problems that we are working on, and more details can be found in the individual group member links.
- Proposing a new class of cooperative communications for multi-node scenarios and characterizing the optimal power allocation for the associated virtual array.
- Proposing cross-layer schemes to answer questions like "who helps whom?" and "how to collaborate?".
- Utilizing possible channel state information at the transmitter to develop more intelligent protocols that decides "When to cooperate?"
- Design distributed space-time coding schemes which takes into account the unreliability of the relay nodes.
- Introducing bandwidth efficient cooperative protocols for multiuser OFDM systems.
- To alleviate the channel estimation burden, we propose differential cooperative protocols and study their performance.
- Propose new distributed relay assignment protocols that take into account the random spatial users distribution. Utilizing the proposed protocols for coverage extension in wireless networks.
- Introducing a novel cooperative multiple access protocol under a TDMA framework. The new proposed protocol has some aspects from cognitive radios, in which the employed relays in the network senses the communication channel and utilizes the empty time slots in order to forward dropped packets from other users. Our study of the proposed cooperative multiple access protocol shows significant increase in the stability region of the arrival rates, in addition to energy efficient savings when optimal relay positioning is taken into consideration.
Dynamic Spectrum Access and Cognitive Radio
Zhu Ji, Ahmed K. Sadek, Charles Pandana, Beibei Wang, Mahmoud Abdulrehem, Yongle Wu, Amr El Sherif
The demand for wireless spectrum use has been growing rapidly with the dramatic development of the mobile telecommunication industry in the last decades. In order to fully utilize the scarce spectrum resources, with the development of cognitive radio technologies, dynamic spectrum access becomes a promising approach to increase the efficiency of spectrum usage, which allows unlicensed wireless users to dynamically access the licensed bands from legacy spectrum holders on a negotiated or an opportunistic basis. Cognitive radio technologies have the potential to provide the wireless devices with various capabilities, such as frequency agility, adaptive modulation, transmit power control and localization. The advances of cognitive radio technologies make more efficient and intensive spectrum access possible.
Our proposed research is focused on developing a game theoretical framework on efficient distributed dynamic spectrum access designed to optimize the spectrum efficiency with the presence of selfish users and minimize the damage caused by malicious users based on the limited local spectrum information, game theoretical optimality criteria and intelligent cognitive engine design. We also perform comprehensive game theoretical optimality analysis and vulnerability analysis of dynamic spectrum access considering the spectrum efficiency, fairness, and major attacks. Moreover, we aim to implement the proposed distributive, efficient and robust dynamic spectrum access through cognitive engine development based on open source cognitive radio platforms.
Broadband and Ultra-Wideband Communications
Wipawee (Pam) Siriwongpairat, Thanongsak (Kee) Himsoon, Quoc Lai
Ultra-wideband (UWB) is an emerging technology that offers great promises to satisfy the growing demand for low cost and high-speed digital wireless home networks. However, the transmitted power level of UWB transmission is strictly limited by the FCC spectral mask to enable peaceful coexistence with other narrowband systems. Due to such restriction, any UWB system faces significant design challenges in achieving the desired performance and coverage range. Our research works to cope with these challenges by developing advanced technologies such as UWB-MIMO, cooperative UWB, and cross-layer design for UWB systems. We also characterize the UWB performance in realistic UWB channel environments which allow us to capture the unique multipath-rich property and multipath-clustering phenomenon of UWB channels.
Cooperative Multimedia Communications
Andres Kwasinski, Hong Zhao, Wan-Yi Lin, Karim Seddik, Ahmed Ibrahim
User cooperation improves communication performance by using the broadcast nature of wireless networks to associate mobile users in such a way that creates a virtual multiple antenna array. The improved performance, revealed as an improvement in received signal quality, comes at the price of a reduced spectral efficiency. When using user cooperation to transmit real-time multimedia signals it is important to find a good tradeoff between the improvement in received signal quality and the reduction in spectral efficiency, as well as a good match with source and channel coding parameters. Other problems of importance in this area are related to the design of the source codec subsystem. Questions such as what is the best coding scheme (single description, multiple description, layered, etc.) and their matching with different user cooperation schemes are also studied in this area.
Cross-Layer Resource Allocation for Wireless Networks
Zhu Han, Andres Kwasinski, Ahmed Sadek, Wipawee (Pam) Siriwongpairat, Thanongsak (Kee) Himsoon, Charles Pandana, Wei Yu, Zhu (Jim) Ji, Beibei Wang
With the advancement of technologies, wireless networks have become ubiquitous, owing to the great demands of pervasive mobile applications. To satisfy the growing requests of wireless services, the future wireless networks are characterized by broadband, high data rate capabilities, integration of services, heterogeneous QoS provisioning, flexibility, and scalability. However, many technical challenges remain to achieve these requirements, such as the adverse nature of wireless channels, scarce wireless resources, and conflicts among users. Resource allocation is an important strategy to combat detrimental effects of wireless channels, optimize the allocations of limited resources, and control the interferences, so as to provide the desired services and optimize the system performances.
The proposed researches aim at developing a unified view on how to efficiently optimize the dynamic allocations of scant wireless resources over assorted wireless network scenarios. These researches are interdisciplinary in that they contain concepts in signal processing, economics, decision theory, optimization, information theory, communications, and networking to address the issues in question. In addition, we provide an innovative insight into vertical integration of wireless networks by cross-layer optimization. Specifically, our current researches focus on dynamic adaptive resource allocations in collaborative transmission networks, ultra wide bade networks, multimedia wireless networks, and ad hoc/sensor networks. There are also some other issues such as network security, MIMO communications, and distributive wireless networking using game theory.
Ad-hoc and Sensor Networks
Charles Pandana, Zhu Han, Wei Yu, Zhu (Jim) Ji, Ahmed Sadek, Wipawee (Pam) Siriwongpairat, Thanongsak (Kee) Himsoon, Karim Seddik
While advances in communication theory and integrated circuit technology have enabled us to build and deploy ad-hoc and sensor networks, how to control, optimize, and effectively collect data from them remains a challenging issue. Many problems addressed in our research group include energy efficient information delivery using cross layer approach, connectivity-aware routing, cooperative routing, routing with data fusion consideration, etc. The solutions of these problems generally can be formulated as nonlinear, large-scale dynamic control problems under uncertainty. In particular, we envision the applications of distributed learning algorithm to tackle these difficult problems. Employing distributed learning algorithm enables us to solve the large-scale complex control problem with very limited information, which is exactly what happened in the practical sensor networks.
Information Assurance and Network Security
Wei Yu, Zhu Ji, Zhu Han, Charles Pandana, Beibei Wang, Mahmoud Abdulrehem
While wireless communication has dramatically changed the way people work and interact, the wireless era continues to be plagued by insufficient security. Due to the unrestricted channel access, attackers can easily launch a variety of attacks ranging from passive eavesdropping from active interfering. It becomes even worse for ad hoc and sensor networks: nodes can be easily captured and compromised due to the lack of enough physical protection. Without necessary countermeasures, even a few compromised nodes can break down the whole network. Traditional cryptography-based security approaches alone are not enough, and necessary monitoring mechanisms have to be launched to detect malicious behavior, such as dropping packets, injecting traffic, and jamming. However, in realistic scenarios, malicious behavior detection could be an extremely challenging task due to the following reasons. First, misbehavior may also be caused by channel noise and uncertainty. Second, in ad hoc and sensor networks, perfect monitoring is either impossible to achieve or too expensive to afford due to the fully distributed nature and limited resources. Both may lead to high false alarm ratio and/or miss detect ratio. Our current research activities focus on establishing secure and reliable communication for wireless ad hoc and sensor networks in noisy and hostile environments based only on local and imperfect monitoring.
Multimedia Forensics and Content Protection
Hong Zhao, Wan-Yi Lin, Steve Tjoa, Matt Stamm
Recent developments in multimedia processing and network technologies have raised the critical issue of protecting multimedia content security and enforcing digital rights. Cryptographic tools (e.g., encryption and authentication) alone are not sufficient since the protection disappears after the data are delivered to users. To address post-delivery protection, multimedia fingerprinting is an emerging technology used to identify users who have legitimate access to the plaintext content but use it for unintended purposes. It provides a proactive means to trace the illegal usage of multimedia by inserting unique identification information (i.e. "fingerprint") into the content before distribution.
Our research has resulted in the design of robust multimedia fingerprints as well as reliable and trustworthy forensic tools for traitor tracing. We also investigate many new challenges in multimedia forensics, such as the behavior dynamics among users in multimedia security systems and how users interact with and respond to each other. This behavioral forensic analysis enables the digital rights enforcer to offer stronger protection of multimedia. Furthermore, for networked multimedia applications, we collectively address security and bandwidth efficiency and design secure fingerprint multicast schemes.
Bioinformatics and Computational Biology
Peng Qiu, Z. Jane Wang
Bioinformatics is an emerging field that combines biology, engineering, etc together. Recently, high throughput measuring techniques (microarray, mass spectrum, etc) have provided massive amount of data, which is believed to reveal biology nature in gene/protein level. However, the data quality is limited by current technologies, and the huge amount of data cannot be understood by visual inspection. Our research work is to apply signal processing methods to improve data quality, and to build models to systematically interpret high throughput experiment data.
For cancer research, we have proposed an ensemble dependence model, which aims at revealing the different dependence patterns between selected features for cancerous samples and noncancerous samples. With the proposed model, we can zoom out to see the big picture, the global dependence relationship among gene/protein clusters; we can also zoom in to examine the details of individual genes/proteins. The model can be used for cancer classification and prediction, biomarker identification.
Biomedical Imaging
Peng Qiu, Z. Jane Wang
PET (positron emission tomography) imaging is a useful tool for image-wide quantification of biochemical and physiologic processes, which is widely used in brain, lung, liver, kidney, etc. Compartment modeling is the most popular approach for PET studies. However, current compartment-model based methods often either require the accurate arterial blood measurements as input function or assume the existence of a reference region. The arterial blood measurement is invasive, painful and risky. In our work, we proposed a subspace intersection idea to estimate the input function, obtaining the PET parametric images in a non-invasive way.