Texas Southern University (TSU) will establish a Center for Research on Complex Networks to conduct research in complex networks. Three research areas of the Center are:

1. Energy Efficient Wireless Sensor Networks (EEWSN), which is to develop energy efficient schemes by integrating secured data aggregation and dynamic power management, investigating efficient multiple adaptation algorithms for uncertainty-aware sensor networks, and exploring dynamic control and optimization strategies;
2. Urban Transportation Environmental Networks (UTEN), which is to develop and implement vehicle emission models and real-time intelligent network control methods to reduce greenhouse gas and pollutant emissions, thereby improving urban air quality;
3. Distributed Computational Networks (DCN), which is to develop and implement algorithms for computing on networks for which the resource allocation is uneven.

The research activities in the Center will involve the disciplines of computer science, mathematics, physics, chemistry, engineering technology, environmental science, and transportation. The research in this Center will be integrated with the science, technology, engineering and mathematics (STEM) education programs. The outcomes of the Center will increase the number of minority and under-represented students who pursue advanced graduate degrees, contributing to meeting the future critical workforce needs of the nation in STEM fields.

The Department of Transportation Studies and the Department of Chemistry will be two of the participating Departments in this subproject of the proposed CREST Center at TSU. The general goal of this subproject is to develop and implement vehicle emission models and real-time intelligent network control methods to reduce greenhouse gas and pollutant emissions, thereby improving air quality in urban transportation environmental networks (UTEN). Three specific research goals and associated objectives will be pursued:

1. To develop and model air quality monitoring networks in an urban environment. The specific research objectives are:
   • Developing fast-response CO2 and NOx electrochemical sensors via intelligent materials design and engineering in selective gas permeable membranes.
   • Optimizing the geographical distribution of air quality monitors.
   • Integrating the gas sensing function into urban transportation environmental networks.
2. To develop vehicle emission models that incorporate real-time transportation activities.
   The specific research objectives are:
   • Developing a Vehicle Specific Power (VSP) distribution databank.
   • Developing VSP-based vehicle emission models for urban networks.
   • Developing a dispersion model for vehicle emissions by incorporating the air quality monitoring data from Goal 1.
3. To develop models that optimize traffic control strategies to mitigate emission and improve air quality on real-time basis. The specific research objectives are: (a) developing models that optimize network topological structures to improve urban air quality; (b) developing models that optimize traffic controls to improve urban air quality; and (c) developing simulation tools that incorporate real-time traffic data and optimization models to support decision-making.

Computer Science and Engineering Technology will be two of the participating Departments in the development of energy-efficient wireless sensor networks (EEWSN) in the proposed CREST at TSU. The goals of the research proposed in this subproject are to develop energy-efficient wireless sensor networks through integrating data aggregation and dynamic power management; to investigate efficient multiple adaptation algorithms for energy-aware sensor networks; and to explore dynamic control and optimization strategies in wireless sensor networks. The specific research goals include:

1. To develop energy-efficient schemes by integrating data aggregation and dynamic power management (DPM), including (a) the modeling and analysis of joint data aggregation and DPM, and (b) investigation of energy consumption and sensor performance.
2. To investigate efficient, multiple-degradation algorithms for uncertainty-aware sensor networks, including (a) discovery of multiple degradation algorithms in sensor nodes, and (b) development of analytical method for energy-efficient whole sensor networks with degradation features;
3. To explore dynamic control and optimization strategies, including (a) optimal design of power mode period and (b) optimization of admission control of data packets.
4. To implement research, including (a) development of efficient simulation tools and efficient application for testing the results of the research, and (b) establishment of modularized wireless communication testing and measurement system.

Remote Lab EEWSN Wiki

Computational approaches for problem solving have proven their value in almost every field of human endeavor. Computers are used for modeling and simulating enumerable complex scientific and engineering problems. From diagnosing medical conditions, controlling industrial equipment, forecasting the weather, managing stock portfolios, and conducting scientific investigations, computers have become essential and irreplaceable tools in these fields. However, one of the limitations of computational simulation science is the speed and memory limitation of the computational hardware utilized. We plan to alleviate this problem by studying methods and means for achieving highly distributed grid computing. Specifically, we will focus on hard computational algorithms that do not traditionally lend themselves to this type of parallelism. Additionally, the general techniques that we will develop within the program will be applicable to many other research endeavors specifically including high performance computing, grid computing, blocking networks and ad hoc wireless networks for which resource allocation is uneven. The goal of this subproject are to improve the efficiency of distributed computing on blocking networks with the integration of mobile ad hoc wireless networks. The specific goals are:

1. To develop distributive computing for computational science: We will develop a general method of distributing the computational workload of complex codes, such as the FreeON code suite and the transportation traffic simulation codes of UTEN.
2. To develop distributed computing for blocking and ad hoc wireless networks; we will develop methods and algorithms for distributed computing on blocking networks, which is highly relevant to EEWSN.
3. To improve the efficiency of communications and protocols for blocking and mobile ad hoc networks.

The Center of Research on Complex Networks (CRCN) project will engage six departments at Texas Southern University in the scholarly activities of the project. These departments are: the Department of Computer Science, the Department of Engineering Technology, the Department of Transportation, the Department of Mathematics, the Department of Chemistry, and the Department of Physics.

Due to the interdisciplinary nature of this project, the facilities used for the three subprojects will be physically housed in different buildings but will be connected. This connection will result in having a single larger facility that ensures the integration as well as the collaboration for all the three subprojects. The CRCN will have the following facilities:

• TSU Energy Efficient Wireless Sensor Networks Lab
• TSU Advanced Transportation Studies Laboratories
• TSU High Performance Computing Center

These facilities will be networked together using secure site-to-site Virtual Private Network (VPN) tunnels over the TSU main campus network. It will ensure that all the resources on these sites work together as if they were located in the same physical location. The design and deployment of the secure VPN tunnels will also include some redundancy to ensure that there is no disruption to the connectivity between the sites. In addition to the site-to-site VPN tunnels, the CRCN will provide the infrastructure for remote users to connect to the facilities via secure remote access VPN connectivity. The following is a diagram that illustrates the integration between the CRCN facilities, followed by a description of the resources included in each facility.

TSU Energy Efficient Wireless Sensor Networks Lab

The Energy Efficient Wireless Sensor Networks Lab will be at state-of-the-art facility for both graduate and undergraduate students participating in the research activities. It will be a remotely accessible laboratory and testbed. The proposed testbed has three sub testbeds: An indoor testbed, outdoor testbed and a body area wireless network testbed. It will be actively used for educational as well as research activities.

• Sensor Nodes: The lab will have a total of 200 iMote sensor nodes which will be used to compose the indoor and outdoor testbeds. The iMote 2 is an advanced wireless sensor node platform built around the low-power PXA271 XScale CPU and integrates an 802.15.4 compliant radio. It has a modular and stackable design with interface connectors for expansion boards on both its top and bottom sides.
• Testbed Server: The server is a MySQL-based database server that will be used to store the experimental data, the information used to generate web content, and the state driving test bed operation. The server will be connected to the lab’s local area network and will be remotely accessible by users through a web interface designed to allow job creation, scheduling, and data collection, as well as an administrative interface to verify testbed control functionality.
• Measurement Instrumentation: It will include the hardware as well as the software needed to take measurements from the sensor nodes. The proposed hardware is based on National Instruments’ NI PXI modular instrumentation and the software is NI LabVIEW which is a powerful graphical environment developed based on the concept of virtual instrumentation (VI) and utilizing computer technologies in combination with flexible software and modular hardware to create interactive computer-based instrumentation solutions. LabVIEW is popularly deployed software used for both academic and industrial applications.

TSU Advanced Transportation Studies Laboratories

TSU has developed advanced transportation laboratories for transportation education and research. Equipment includes an Autoscope Mobile Traffic Van, Full-Motion driver Simulator, real-time traffic monitoring system, and a portable emission measurement system (PEMS). TSU has signed an agreement with TxDOT in Houston TranStar by which TSU is able to access the real-time traffic data provided by TranStar. TSU also owns additional equipment, such as GPS, traffic counters, etc. All of this advanced equipment will provide a solid infrastructural foundation to support the research as well as the education delivery proposed for this project. In addition, all of the researchers including graduate research assistants in the Department of Transportation Studies will be have state-of-the-art personal computers. A computer laboratory is available in the department, which includes 20 networked workstations. All computers are networked and connected to the Internet through TSU’s mainframe computer network system, which enhances the collection and dissemination of research information. The TSU central library is a member of the Houston Area Research Library Consortium (HARLC), which includes Rice University, the University of Houston, Praire View A&M University, Texas A&M University, and the University of Texas.

Currently, the computers in the Department of Transportation Studies are installed with the following transportation software: CORSIM, VISSIM, INTEGRATION, TRANSYT-7F, PASSER, ArcView, TransCAD, EMME/2, QRS II, MOBILE6, EMFAC, CMEM, ONROAD, STNCHRO, DYNAMIC, etc. In addition, these computers have most of the general computer application software installed, such as MS Office, statistical analysis tools (SPSS), and others. The installed computer programming compilers include Power FORTRAN, Visual BASIC, Visual C++, MATLAB, etc.

TSU High Performance Computing Center

Texas Southern University's High Performance Computing Center (TSU-HPCC) was established in 2008 to promote research and teaching on campus through integrating leading edge high performance computing and visualization for the faculty, staff, and students of Texas Southern University. The HPCC provides consulting and assistance to campus researchers who have experimental software and/or hardware needs. We also provide training in parallel and grid computing. HPCC will serve as a liaison between the various teams engaged in the proposed research. We work to support, configure, and port applications to HPCC resources (http://hpcc.tsu.edu).

• Ares, installed in December 2008 has sixteen dual-slot, quad-core nodes with Intel Xeon 5350 2.0 GHz processors with 8 Gigabytes of memory connected via dual Gigabit Ethernets. The full parallel cluster has a total of 128 cores and a total memory of 128 Gigabytes, with a peak speed of 0.672 Teraflops (http://ares.tsu.edu/).
• Hades, installed in January 2010, has eight dual- slot, hyper- threaded quad core nodes with the Intel E5520 2.33 GHz Xeon Processor with 12 Gigabytes of Memory connected via a 10 Gigabit Ethernet using an ultra low latency Arista 7124S switch. The full parallel cluster has a total of 128 virtual cores and a total memory of 96 Gigabytes, with a peak speed of 0.783 Teraflops.
• Charon, installed in February of 2010, is a Condor (OpenMPI configuration) grid- based distributed computer system. At present, it has eight 3.0 GHz Intel xenon’s with 1GByte of memory, four duel- core 2.5 GHz with 2GByte of memory and two quad- core 2.2 GHz with 4 GByte of memory. The Ethernet is variable between 100 MBit to 1 GBit. It will be our main test bed for development of the ideas within this proposal on grid- based and blocking networking.

Except for the equipment we propose to purchase, the following equipment in the Computer Labs of the Computer Science and Engineering Technology Departments will be used by all the CREST teams:

• Eight NI PCI-5112,100 MHz, 100 MS/s 8-Bit Digitizer.
• Seven NI PCI-6024E, 200 kS/s, Multifunction DAQ.
• One NI SC-2075, connector board.
• Ten NI LabVIEW 8.5.
• Six multi-channel oscilloscopes.
• Five Mathworks Matlab 7.0.
• One HP ProLiant DL365 server
• Seven NI ELVIS.