De La Salle University’s Advanced Research Institute for Informatics, Computing and Networking (AdRIC) hosted a hybrid lecture on August 3, 2026, featuring Dr. Dominic Kitavi, Senior Lecturer at the Department of Mathematics and Statistics, University of Embu, Kenya, and Senior Postdoctoral Research Fellow at De La Salle University.
Entitled “Developing High-Impact Research: A Case Study in Statistical Signal Processing,” the seminar explored how researchers can move from curiosity and ideas toward research that addresses meaningful problems and contributes knowledge that others can build upon.
Dr. Kitavi began by asking three fundamental questions: Why do we conduct research? What makes some research more impactful than others? And how can researchers develop work that makes a difference?
Rather than viewing research simply as a way to produce more publications, Dr. Kitavi emphasized that meaningful research should aim to solve important problems and create knowledge that can be useful to others. He highlighted several characteristics of high-impact research, including novelty, rigor, significance, reproducibility, practical relevance, and generalizability.
He explained that impactful research is built through a series of connected steps, beginning with a real-world problem and literature review, followed by identifying a research gap, developing research questions, selecting appropriate methods, conducting analysis and validation, and communicating the findings through publication. Each stage contributes to the overall strength and value of the research.
Dr. Kitavi also discussed where research ideas can come from. Literature gaps, real-world problems, new technologies, existing theories, interdisciplinary collaboration, and curiosity can all serve as starting points for research. He emphasized that some of the strongest research ideas emerge when important problems meet genuine curiosity and careful thinking.
To demonstrate how these principles can be applied, Dr. Kitavi presented a case study in statistical signal processing, a field concerned with extracting useful information from uncertain and noisy data using mathematical and statistical methods. He discussed applications of this field in areas such as wireless communications, radar, acoustic sensing, medical imaging, and other technologies that rely on accurate information from measurements.
The case study focused on the challenge of estimating the direction of a signal source when measurements are affected by noise and the characteristics of the sensors are not completely known. Dr. Kitavi walked the audience through how he identified a research gap in existing directional sensor models and developed research questions around creating a more general model that could estimate both the source direction and sensor directivity.
From there, he demonstrated how research questions can guide the development of a research framework. The study involved developing a generalized directional sensor model, describing how noisy measurements relate to unknown parameters, and examining whether those parameters could be reliably estimated. He introduced tools such as the Fisher Information Matrix and Cramér-Rao Bound to evaluate the information contained in the observations and the limits of estimation accuracy.
One of the key lessons from the case study was that mathematical tools should support the research questions rather than determine them. Dr. Kitavi emphasized the importance of first understanding the problem and identifying what needs to be answered before deciding which methods should be used.
He also discussed the role of numerical experiments in strengthening research findings. Rather than conducting experiments simply to generate figures, researchers should design them to answer specific questions and test whether their theoretical expectations hold. In his case study, factors such as signal-to-noise ratio, number of observations, sensor directivity, and source direction were examined to understand their effects on estimation accuracy.
The results demonstrated, among other findings, that estimation accuracy improves as the signal-to-noise ratio increases and as more observations become available. The numerical results also supported the theoretical analysis, illustrating how simulations can provide evidence for scientific claims.
Beyond the technical aspects of the case study, Dr. Kitavi also shared practical advice on communicating and publishing research. He encouraged researchers to write with their readers in mind, clearly distinguish their contribution from previous work, and ensure that every section of a paper serves a purpose. He also reminded participants that figures should help explain the findings rather than simply decorate the paper, and that developing a strong manuscript requires repeated revision.
The seminar also covered the importance of choosing an appropriate journal. Dr. Kitavi encouraged researchers to consider factors such as the scope and audience of a journal, manuscript length, expected level of novelty, publication model, and review time. He emphasized that journal selection should be part of the research strategy rather than something considered only after the research has been completed.
Discussing the peer-review process, Dr. Kitavi reminded participants that rejection is a normal part of research. Reviews and editorial decisions can provide opportunities to identify weaknesses, improve a manuscript, and strengthen the research. Persistence, he noted, is an important part of becoming a successful researcher.
Dr. Kitavi concluded the seminar by returning to the central message of his talk: high-impact research is not a matter of luck. It is developed through curiosity, critical thinking, rigorous methodology, and effective communication. Researchers can begin by identifying an important problem, reading the literature critically, defining a clear research gap, asking focused questions, developing a sound methodology, and communicating their findings clearly.
For the graduate students and researchers who participated in the hybrid seminar, the discussion provided both a closer look at statistical signal processing and practical lessons that can be applied to research across different fields. More importantly, it encouraged participants to think beyond simply completing a study or publishing a paper and instead consider how their research can address meaningful questions and create knowledge that others can build upon.
The seminar served as another opportunity for DLSU’s research community to learn from the experiences of researchers working across different disciplines and countries, while encouraging students and emerging researchers to develop research that is rigorous, relevant, and meaningful.