I launched two CS workshops for young girls in fourth and fifth grade as part of a larger Girl Scouts of America initiative. These workshops embrace playfulness as a fundamental design constraint to ensure an early positive association with STEM concepts. Students are exposed to basic coding skills through simple robots that are programmed to play games. Survey results show that these workshops lead to a stronger interest in CS for the participants.
Paper

Growing interest in CS has led to a shortage of qualified educators for K-12 level instruction. Many out-of-school-time computing programs are led by well-intentioned individuals with little formal training in education such as undergraduate students pursuing computing degrees. This results in environments that are often not conducive to learning for students coming from diverse cultural and educational backgrounds. I addressed some of these challenges by using undergraduates as coaches in K-12 outreach CS activities for underrepresented students. These coaches are given essential pedagogical training with playful material that allows them to succeed as educators.
Paper

I developed two mobile station placement algorithms to meet the critical communication requirements of first responders in an ad hoc public safety network. By considering the class of first responders and user equipment applications, I provide an efficient base station placement algorithm to maximize critical communication needs according to priority levels.
Paper

I defined different classes of mobile base stations that have varying performance characteristics and devised three different first responders mobility models. My proposed algorithm applies the modern clustering technique to deal with the characteristics of different kinds of mobile base stations.
Paper

I generated two channel assignment algorithms (centralized and distributed) with considering switching overhead. Switching overhead is a major factor that limits overall network throughput in dynamic channel assignment schemes. I extended existing algorithms, Greedy Maximal Scheduling and Distributed Maximal Scheduling, taking the switching delay into account in the channel assignment.
Paper Paper

I developed a RL based queue management scheme that supports QoS in multi-radio multichannel wireless mesh networks. I developed an algorithm that takes into account differentiated queue management and channel assignment together. I used the technique of RL for the important step of assigning packet to one of the queues.
Paper

By applying reinforcement learning (RL)-based queue management and scheduling methods, I designed an intelligent patient data transmission scheme in Wireless Hospital Networks. In this scheme, I used a game-theoretical approach where patients compete for transmission resources by assigning different utility values to data packets. These utility functions are largely based on data criticality and deadline, which together determine the data’s scheduling priority.
Paper

I established an on-line math education system that provide higher quality education by allowing the users to work through the curricula at a self-set pace. I developed a parser based on context-free grammar to generate the semantic meaning of each mathematical problem at a high-school level.

I implemented a decision support system that optimizes service-network combinations within multi-access wireless network environments considering network load and service traffic characteristics both from marketing and network engineering points of view.
Paper

I built a large-scale wireless network simulator that analyzes the performance of Common Radio Resource Management (CRRM). I used a utility-based approach to access selection in Multi-access Wireless Networks.

I created a QoS-aware policy-driven scheduling algorithm in HSDPA (High-Speed Downlink Packet Access). I considered the problem of scheduling and rate assignment on the forward link while maximizing total utility in a multiuser system where multiple users can be scheduled during each interval. I developed a knapsack-like greedy algorithm to the utility maximizing problem and proved that the total utility under the algorithm converges to the maximum.
Paper1 Paper2

I constructed various uplink scheduling algorithms for HSUPA/EvDO Rev. A/WiMAX. Utility function based Common Radio Resource Management algorithms, Round Robin, Proportional Fair, and Maximum C/I algorithm were developed and implemented in UGC Service Network Test-bed.
Paper1 Paper2