Passionate about practical solutions to complex problems, Talmai holds a B.Sc. in computer science, a M.Sc. in mechatronics and a Ph.D. in computer science and engineering. If he were to describe his research focuses, it would probably include dependable heterogeneous (wireless) networks, as well as dealing with inherently massive uncertainty issues. He is also very fond of fundamental group communication primitives.

Papers and Research

University of Cincinnati

PhD, Computer Science and Engineering
2008 – 2012

GPA: 3.994/4.000

Universidade Federal da Bahia

MsC, Mechatronics
2003 – 2007

GPA: 8.59/10.00

Universidade Federal da Bahia

BsC, Computer Science
1998 – 2002

GPA: 9.25/10.00

Significant Publications

  1. The Internet of Things is poised to transform lighting from a simple illumination source, which is most often taken for granted, into a smart and data-rich infrastructure for the cities.

    [1] A. Murthy, Y. Yang, D. Jiang, T. Oliveira and D. Cavalcanti. Internet-of-Things Enables a Cyber-Physical Systems Approach to Lighting. Workshop on Big Data Analytics in CPS: Enabling the Move From IoT to Real-Time Control, held in conjunction with CPS Week 2015. April, 2015.

    [2] A. Murthy, D. Han, D. Jiang and T. Oliveira. Lighting-Enabled Smart City Applications Ecosystems Based on the IoT. Accepted for publication at the 2015 IEEE 2nd World Forum on Internet of Things (WF-IoT). December, 2015. Milan, Italy.

  2. Research and development on dependable heterogeneous wireless networks (HWN) and dealing with the inherently massive uncertainty issues. Inspired by challenges of multi-constraint path selection and the need for providing a desired QoS, worked on devising an efficient network selection algorithm that applies fuzzy logic, minimization of maximal regret (MMR) and other multiple attribute decision making (MADM) techniques. Considered both dynamically changing network conditions, as well as user uncertainty. Exploiting underlying characteristics of a HWN, proposed solution(s) balance the situation with an intelligent and robust decision making algorithm that is capable of efficiently selecting the best network in forwarding the data. Software used for simulations: MATLAB, NS-2 and simulator implemented in C++.

    [1] T. Oliveira, S. Mahadevan and D. P. Agrawal. Handling Network Uncertainty in Heterogeneous Wireless Networks. In proceedings of 2011 IEEE INFOCOM, pg. 2390-2398. ISBN: 978-1-4244-9921-2. (Acceptance rate: 15.96%).

    [2] T. Oliveira, M. Raju and D. P. Agrawal. Accurate Distance Estimation Using Fuzzy based combined RSSI/LQI Values in an Indoor Scenario: Experimental Verification. Network Protocols and Algorithms Journal, Special Issue on 2nd IEEE Smart Communication Protocols and Algorithms, vol. 4, No. 4, pages 174-199, 2012. DOI: 10.5296/npa.v4i4.2173.

    [3] T. Oliveira, S. Mahadevan and D. P. Agrawal. Using Minimization of Maximal Regret for Interface Selection in a Heterogeneous Wireless Network. Accepted for publication at the 2nd IEEE International Workshop on Smart Communication Protocols and Algorithms – part of ICC 2012. 2012.

    [4] M. Raju, T. Oliveira and D. P. Agrawal. A practical distance estimator through distributed RSSI/LQI processing – An experimental study. Accepted for publication at the 2nd IEEE International Workshop on Smart Communication Protocols and Algorithms – part of ICC 2012. 2012.

  3. While results are encouraging, no optimal strategy has yet been identified that could estimate link quality and incorporate both the link reliability measurements as well as the bandwidth capacity in Wireless Mesh Networks (WMN). Furthermore, link estimation remains an open problem. Considering multi-user environment, any optimal solution would also need to consider multiple communication flows. Explored possible metrics to use as link-attribute (LQ) estimators, and formulated one that is capable of identifying the quality of communication between neighboring mobile devices while maintaining scalability. Proposed an approximate dynamic programming (DP) solution capable of utilizing limited network knowledge and stochastic process. Then examined the trade-off between reliability and energy usage since in practice it becomes very hard to determine when to stop retransmitting a message in order to ensure proper delivery. By relying on the proposed link-quality estimator, a maximum number of attempts is computed which (probabilistically) ensures delivery while maintaining an energy-efficient network. Through appropriate adjustments, the estimator is capable of obtaining the bounds on the number of retransmission attempts required to increase the message delivery ratio. Analytical results mainly based on probability theory and stochastic dynamic programming with approximation. Software used for simulations: MATLAB, NS-2 and simulator implemented in C++.

    [5] T. Oliveira and D. P. Agrawal. Stochastic Approximate Dynamic Programming with Link Estimation for High Quality Path Selection in Wireless Mesh Networks. In Workshop on Complex and Communication Networks (CC- NET 2010) in conjunction with the IEEE Globecom 2010, Miami, Florida, USA. p. 430-434, 2010

    [6] T. Oliveira, et. al. Reliability Tradeoffs for Energy Efficient Wireless Networks. Accepted for publication in the IEEE International Workshop on Mobile Ad-Hoc Wireless Networks iWMANET 2012, conjointly organized with iCOST 2012 in Avignon/France. 2012.

  4. Through continuous interaction with interdisciplinary groups throughout the University of Cincinnati developed a wearable sensor network for ballet dancers. In this project, by relying on fuzzy logic, implemented a gesture detection algorithm that runs on a telosb tmote and uses a single accelerometer. Tested on over 30 different dancers and students, the algorithm captures most of the hand locations with minimal errors.

    [7] M. Helmuth, et. al., Water Birds: Compositional Collaboration with Clarinets, Wireless Sensors, and RTcmix. Accepted for publication in the Journal of the Society for Electroacoustic Music, and recommended for inclusion in SEAMUS 2011. 2011

    [8] M. Helmuth, et. al. Wireless Sensor Networks and Computer Music, Dance and Installation Implementations, In Proceedings of 2010 International Computer Music Conference, New York, NY. 2010.

    [9] M. Helmuth. et. al., Wireless Sensor Networks and Percussion Tracking: Ocean Drum. Accepted for publication at the International Computer Music Conference (ICMC) 2012, demonstration track. 2012.

  5. Explored handover issues in heterogeneous wireless network (HWN), WiMAX femtocells and vehicular ad hoc networks (VANET). In HWNs, load balancing was obtained by introducing an effective traffic load balancing scheme for maximizing expected throughput. Analytical results were based on queuing theory, and two scenarios were explored: one where multiple APs of a HWN were modelled as a cascade of in- dependent M/M/1 queues, and the second where they were modelled as a union of multiple M/M/1 and M/D/1 queues. The expected number of packets in the system is formulated as a convex optimization problem and solved using a Lagrange multiplier. A recursive approach was used in distributing the traffic to different APs. In the latter model, the time spent by each job in the system is minimized by using a solver to obtain optimal rate distribution among the networks. The work with WiMAX femtocells noticed that conventional handover methods cannot support a good enough performance under different mobility patterns and dynamic network conditions. An appropriate call admission control was proposed, and a resource management scheme was then introduced that significantly reduce unnecessary handovers. Finally, with regards to VANETs, seamless connectivity is addressed through a novel hybrid communication paradigm where both the existing network infrastructure through a Vehicle-to-Infrastructure (V2I) protocol, as well as a traditional Vehicle-to-Vehicle (V2V) protocol are explored in order to achieve QoS. A load balancing mechanism is also proposed that decreases the message propagation delay of high priority packets regardless of the network overload. Simulations were done on NS-2, implemented in C as well as in Java.

    [10] K. Louha, J. H. Jun, T. B. Oliveira and D. P. Agrawal, Exploring Load Balancing in Heterogeneous Networks by Rate Distribution. International Journal of Autonomous and Adaptive Communications Systems (IJAACS). Vol. 3, No. 3, pp 284-307, 2010.

    [11] R. Singoria, T. Oliveira and D. P. Agrawal. Reducing Unnecessary Handovers: Call Admission Control Mechanism for Handover between WiMAX Macrocells and Femtocells. Accepted for publication at the IEEE Globecom 2011. 2011. (Acceptance Rate 36.61%)

    [12] A. Mostafa, et. al. A V2X-based approach for reduction of delay propagation in Vehicular Ad-Hoc Networks. In Proceeding of the International Workshop on Seamless Connectivity in Vehicular Networks (SCVN 2011), part of the 11th IEEE International Conference on ITS Telecommunications. 2011.

    [13] A. Mostafa, et. al. QoSHVCP: Hybrid Vehicular Communications Protocol with QoS Priorization for Safety Applications. ISRN Communications and Networking, vol. 2012, Article ID 149505, 14 pages, 2012. doi:10.5402/2012/149505.