Show positions, publications and grants

Bilal Khan: positions, publications and grants

Bilal Khan is Professor of Data Science, Professor of Computer Science and Director, Health Data Warehouse at Lehigh University. The interactive map above follows 148 publications in 18 research topics (cybersecurity and digital communications, public health and sociology, foundations) from 1998 to 2026, with 29 funded grants and 53 students and postdocs mentored (6 postdocs, 10 Ph.D. students, 15 master’s students). Download the CV (PDF).

Positions

Lehigh University
- Professor of Data Science in the Dept of Biostatistics & Health Data Science Jan. 2022 – present
- Professor in Dept of Computer Science Jan. 2022 – present
- Director, Health Data Warehouse Jan. 2022 – present
University of Nebraska-Lincoln
- Adjunct Faculty, Department of Sociology Jan. 2020 – present
- Co-Director, Longitudinal Networks Core, Rural Drug Addiction Center (NIH CoBRE) July 2019 – Jan. 2020
- Happold Professor of Computer Science and Sociology Jan. 2016 – Jan. 2020
John Jay College of Criminal Justice, City University of New York
- Research Affiliate, Center for Drug Use & HIV Research (NYU) Aug. 2011 – Aug. 2019
- Professor of Mathematics and Computer Science Sep. 2009 – Dec. 2015
- Co-Director, NYC Social Networks Research Group Jan. 2009 – Dec. 2015
- Doctoral Faculty in Criminal Justice Jan. 2008 – Dec. 2015
- Doctoral Faculty in Computer Science Jan. 2005 – Dec. 2015
- Director of Laboratory for Cybersecurity Sep. 2004 – Dec. 2015
- Research Affiliate, Center for Court Innovation (NYC) Oct. 2009 – Mar. 2012
- Associate Professor of Mathematics and Computer Science Sep. 2004 – Aug. 2009
Center for Comp Science, U.S. Naval Research Laboratory
- Research Scientist, Networks Group 1996 – Sep. 2004

Education

Ph.D. in Mathematics, City University of New York, 2003 (advisors: A. Miasnikov, J. Pach, J. Roitberg) – Algorithmic problems on automorphic conjugacy in the free group of rank 2.
M.Sc. in Computer Science, Johns Hopkins University, 1997 (advisor: B. Awerbuch) – New algorithms for topology aggregation and competitive on-line routing.
B.Sc. in Mathematics & Computer Science, Massachusetts Institute of Technology, 1993 (advisor: M. Sipser)

Research areas

Graph theory, combinatorics & graph arithmetic

Mathematical work on graphs and their structure. It defines metrics and distances between graphs, develops an arithmetic of graphs (sums, factorization and irreducibility), and studies related combinatorial objects such as Goldbach sets and polytopal uncertainty. A recurring practical aim is scalable network design: building physical topologies that support overlays which compress or expand the distances between nodes. (9 publications, 2003–2014)

Combinatorial group theory & cryptography

Algorithmic and structural problems in combinatorial group theory, centred on free groups. The Ph.D. work analysed the structure of automorphic conjugacy in the free group of rank two; related papers address the Hanna Neumann conjecture on subgroup intersections and positive theories of groups, and part of this work was NSA-funded. A cryptographic strand proposes non-commutative key-exchange schemes built on polycyclic groups, and recent work studies the geometry of automorphism orbit graphs in rank two. (5 publications, 2002–2026)

Other mathematics & physics (music metrics, cellular automata, Casimir)

A set of smaller mathematical projects outside the main research lines. They include similarity measures and metric spaces for models of musical tonal hierarchy, cellular automata and Boolean networks used to study morphogenesis, thermal robustness and obstructions to growth in formal dynamical systems, and two 2026 preprints that represent the Casimir trace for parallel-plate geometries as quadratic forms. (7 publications, 2006–2026)

ATM / PNNI routing & signaling

Routing and signaling in ATM networks, the subject of the M.Sc. thesis work. The central problem is routing through hierarchical networks whose domains reveal only an aggregated summary of their internal structure (PNNI topology aggregation), and how to design such networks well. The work also produced SEAN, an open ATM signaling implementation, and PRouST, a routing and simulation toolkit, so that these protocols could be studied experimentally. (7 publications, 1998–2002)

Network simulation toolkits (CASiNO, SEAN, PRouST, CHIME)

Software frameworks for building and studying network protocols and simulators. CASiNO is a component architecture for simulating network objects, with a modular dataflow design and a reactor kernel; SEAN and PRouST support ATM signaling and PNNI routing research; CHIME is an information-modeling environment in which information is organised into units called cells. The goal was to make protocol implementation, simulation and experimentation fast, reusable and open. (3 publications, 1999–2002)

Optical (WDM) network routing

Control and routing for all-optical wavelength-division-multiplexing (WDM) networks. The work lays out the algorithmic problems that a routing protocol for lightpaths must solve, describes the optical network control plane, and measures how much advertising wavelength availability improves wavelength selection. TRON, a toolkit for routing in optical networks, supports experiments with these protocols. (4 publications, 2000–2004)

Agent-based network management & multi-agent systems

Network management by distributed, cooperating software agents. OPTIPRISM is a hierarchy of mobile agents that provides configuration and fault management for all-optical networks; companion papers pose the mathematical problems of deploying such hierarchies at minimum communication cost, balancing load among agents, and building trust and accountability into multi-agent systems. Later work predicts failures in autonomic networks by trend analysis across several time scales. (7 publications, 2000–2013)

Mobile ad hoc & wireless networks

Keeping mobile ad hoc and wireless networks connected and efficient under tight energy limits. The papers allocate a fixed power budget among nodes to extend connection lifetime or to reduce bit and packet errors, optimise routes (variants of AODV) to cut control traffic, and use cooperative mobility, moving nodes deliberately to support quality of service. Emergency-response teams were a motivating application, supported by NSA funding. (20 publications, 2006–2016)

Hybrid RF / free-space-optical networks

Hybrid radio-frequency and free-space-optical (RF/FSO) networks, which combine two media so that each offsets the other's weaknesses. The work designs topology-control schemes for better quality of service in large hybrid mesh networks, including distributed GPU-based algorithms, and a game-theoretic mechanism that uses virtual payments to guarantee that nodes cooperate. (4 publications, 2009–2012)

Network security, forensics & malware

Detecting, tracing and defending against attacks on networks and hosts. Topics include peer-to-peer worm detection and immunization, an agent-based system that reconstructs malicious flows to find the sources of DDoS attacks, securing BGP, covert channels built from packet permutations, and memory confidentiality for legacy software. Later work asks how malware evolves, measuring its biodiversity and relationships through static analysis. (13 publications, 2007–2015)

Cognitive radio & bio-social spectrum access

Dynamic spectrum access in cognitive-radio networks, where unlicensed secondary users opportunistically use spectrum left idle by licensed users. The distinctive idea is bio-social: treating secondary users as foraging organisms with social behaviours such as deference, hunger and pecking order, and using evolutionary game theory to anticipate how societies of such devices organise and to design secure, fair access schemes. NSF funded the behavioural-ecological network models behind it. (9 publications, 2014–2021)

Social networks in Indigenous communities (Labrador, Alaska)

Social-network studies of Indigenous communities, chiefly Labrador Inuit in Nain and Alaska Native communities. They examine kinship, family and exchange, food sharing and subsistence, population movements after land claims, and the structure of helping networks. A methodological aim is a way to acquire perceptual (cognitive) network data from multiple reports on randomly chosen pairs of people. The work was funded by NSF and NIH. (7 publications, 2011–2018)

Hidden populations: network sampling, estimation & criminal-justice research

Methods and studies for populations that are hard to reach. The methods side covers respondent-driven and network sampling and estimators of hidden-population size (network scale-up, capture-recapture). They are applied in criminal-justice research: the retail methamphetamine market in New York City, the underground commercial sex economy in eight US cities, the commercial sexual exploitation of children, and victims of domestic minor sex trafficking, largely funded by NIJ and OJJDP. NSF-funded research-experience programmes trained students in social network analysis. (12 publications, 2008–2018)

HIV / HCV risk networks among people who inject drugs

How the structure of social and risk networks among people who inject drugs shapes the spread of HIV and hepatitis C. The studies cover network topologies, needle sharing and partner restriction, social capital and stigma, and HIV/HCV co-infection, in New York, rural Puerto Rico and the US Midwest. The aim is to inform prevention and treatment for these populations, with NIH funding. (14 publications, 2011–2020)

Substance use & mental-health behavior studies

Close, in-the-moment study of substance use and emotional regulation using the ODIN app for ecological momentary assessment. The work tests whether daily data collection is feasible with people who use drugs or are in residential treatment, and examines how craving, affect, stress and alcohol use co-evolve, including alcohol myopia, PTSD symptoms and experiential avoidance. It aims to identify predictors of return to use and of dropping out of treatment. (5 publications, 2022–2025)

Network modeling & simulation (epidemic and health systems)

Modeling and simulation of networked systems, with emphasis on disease and health care. Stochastic agent-based simulations show how risk-network structure shaped the spread of HIV (a 'firewall effect') and test combination interventions and the cost-effectiveness of treatment for hepatitis C; others model pathogen propagation, caregiver scheduling and device heterogeneity in critical care, and, more recently, avian influenza and COVID-19 forecasting. The book Network Modeling and Simulation: A Practical Perspective gives the methodological foundation. (10 publications, 2010–2025)

Data science & machine learning for behavior data

Machine learning for longitudinal human behavior. The ODIN platform collects continuous-time data on personal change and interpersonal interaction; models built on it predict short-term substance-use behaviors, augment limited tabular data (GAN-based) and generalise across data distributions (ConText-LE, MuHBoost). The aim is just-in-time intervention for at-risk individuals. (5 publications, 2019–2025)

AI & language models for learning and text

Applying machine learning and large language models to education and text. The work forecasts students' academic trajectories and engagement in STEM courses from longitudinal experiential data using pretrained language models, tests just-in-time AI-generated interventions in a randomised trial, clusters trajectories to assess adaptive interventions, and includes a book on prompt engineering. NSF programmes support it. (7 publications, 2023–2025)

Publications