The University of Arizona is renewing its participation in the Connection One (C1) center, an I/UCRC center that was created in 2002. The lead institution is Arizona State University, and the center at present includes five universities and more than twenty industry members. The main research mission of the C1 is to develop technologies and solutions for emerging wireless communication systems, ranging from circuit designs and smart antennas to wireless network architectures and protocols. The scope of C1 extends to the integration of wireless and broadband wire-line technologies (optical communications).
The primary focus of the proposed site over the next five years will be to increase the scope of the present work focused on communication protocols for wireless systems as well as mixed analog/digital circuit designs. The extensions to areas of research in wireless technology such as security and RFID are very positive. The cognitive radio area is also an important area for contribution as well as the integrated sensor area.
The activities proposed by the University of Arizona (UA) research site will impact many important technology needs in the commercial and public sectors. The use of industrial internships to enhance the educational experience is very attractive. The site director at UA plans to exploit the NSF SBIR/STTR program to allow small businesses to directly participate in the C1?s research, and anticipates at least two new small companies to be part of the joint UA/industry projects. The site director has a diversity plan for the I/UCRC that makes specific efforts to attract under-represented groups to fill research, undergraduate and graduate positions.
Project Outcomes Connection One (C1) was established in 2002 as an NSF Industry University Cooperative Research Center (IUCRC). The University of Arizona (UA) joined C1 in 2003, as a center site. In subsequent years, three other sites jointed C1, with a total of more than 26 affiliates from industry and government labs. By 2008 the UA completed Phase I of its participation in C1 and started a 5-year Phase II. The current NSF award and reported outcomes pertain to Phase II. C1’s mission is to develop technologies that enable end-to-end reliable and secure communication for a variety of applications. Its research agenda includes designing next-generation reconfigurable antennas, low-power IC chips, advanced transistor models, and wireless system designs and protocols. During Phase II, the research projects for the UA site focused on dynamic spectrum access (DSA), wireless security, low-power wireless devices, wireless sensors, and multi-antenna (MIMO) systems. Examples of some of the funded projects and their key outcomes are as follows: Project 1: Interference-aware Routing and Reservation Mechanisms for Multi-Hop Cognitive Radio Networks The main objective of this project was to develop efficient routing mechanisms for a mobile ad hoc network of cognitive radios (CRs). The investigators designed channel-adaptive path selection algorithms that optimize a network-wide objective function, including the spatial network throughput and the total energy consumption, subject to power, spectrum, and rate constraints. On-demand and proactive mechanisms were developed for distributed implementation of the route discovery process. Cross-layer solutions were also investigated. Project 2: Securing Spectrum Access in Cognitive Radio Networks In this project, the investigators studied the impact of sophisticated adversaries that jam the common control channel (CC) in a DSA system, with the goal of preventing CR users from performing any collaborative function, such as spectrum sensing, multiple channel access, etc. These attacks are particularly devastating for CRNs, since they eliminate the frequency agility advantage inherent to their SDR architecture. Randomized CC negotiation protocols were developed, which increase the uncertainty of an adversary with respect to the "location" of the CC. These protocols allow the identification of any compromised CR nodes (insiders) that facilitate the adversary’s goals. Project 3: Efficient Mechanisms for Sequential Channel Sensing/Probing in Practical CR Systems under Sensing Inaccuracies When a secondary CR device wishes to opportunistically access the spectrum, it must first scan (sense) channels sequentially and select the ones that are deemed idle. The sensing process can be done passively or actively. Passive sensing relies only on the state of the channel as seen by the CR transmitter, and considers this state to be representative of channel availability at the receiver. Despite its simplicity, this approach overlooks heterogeneity in spectrum availability and channel asymmetry due to multi-path fading. In this project, optimal design and operation of active sensing was studied, with the goal of maximizing the system throughput. A throughput-optimal joint sensing/probing scheme was designed for practical CR systems. This scheme accounts for the instantaneous variations in channel conditions as well as realistic sensing errors. Project 4: Interference and Jamming Mitigation in Satellite Communications Using Spectrum Sensing and Dynamic Frequency Hopping A novel application of CRs in the context of satellite communications was investigated. Satellite systems are prone to unintentional as well as intentional (jamming) interference. Traditional anti-jamming spread-spectrum techniques employ a secret pseudorandom-noise sequence. Although this approach works well in random interference scenarios, it performs poorly against smart jammers. The approach taken in this project relies on exploiting the spectrum sensing capabilities of CRs for proactive detection of interference. A CR module simultaneously monitors a sliding window of frequencies that the transmitter will soon start hopping over according to its fixed FH sequence. Depending on observed interference, the CR decides which of monitored frequencies are to be skipped. By focusing on "future" frequencies, the proposed approach prevents disruptions to the ongoing communications. Project 5: Protocols for Virtual MIMO in Multi-hop Wireless Sensor Networks This project explored a novel approach for improving the energy efficiency of a wireless sensor network (WSN) via virtual multi-input multi-output (VMIMO) techniques. In VMIMO, a group of single-antenna nodes forms a virtual multi-antenna transmitter/receiver, exploiting the spatial diversity to reduce the required total energy. Distributed clustering techniques were developed, which enable sensor nodes to group themselves into virtual multi-antenna transmitters and receivers. These techniques account for the residual energy and battery discharge characteristics of various nodes (so as to maximize the operational lifetime of the WSN) as well as the total power consumption along a multi-hop VMIMO route. This NSF award contributed to the education of many PhD and MS students, as well as several undergraduate students. Hundreds of journal articles and peer-reviewed conference papers were published, along with several invention disclosures and pending patents. Results from this project were disseminated to the public via publications, invited presentations and keynotes, and briefings to industry affiliates.
