Guneesh Vats supervised by Dr. Arti Yardi received his Master of Science – Dual Degree in Electronics & Communication Engineering (LED) by Research. Here’s a summary of his research work on Change-Point Detection via Structured Projections: From Channel Codes to Embeddings

Change-point detection (CPD) is the problem of detecting abrupt changes in the statistical properties of a sequence of observations. Given a sequence whose underlying distribution changes at an unknown time instantcalled the change-pointthe goal is to estimate when this change occurred. CPD has a rich theoretical foundation, with applications spanning quality control, nancial time series analysis, climate monitoring, and biomedical signal processing.

This thesis studies the CPD problem in the context of channel-coded and modulated communication systems. In modern wireless communication, adaptive modulation and coding (AMC) is employed to improve spectral eciency: the transmitter dynamically adjusts its channel code and modulation scheme based on the prevailing channel conditions. For the receiver to correctly decode the transmitted data, it must identify the transmission parameters in use. While blind identication methods can estimate these parameters directly from the received noise-aected data, they assume that the parameters remain xed throughout the observation. In practice, the transmitter may switch its parameters at an unknown time instant, violating this assumption. It is therefore essential to rst detect the change-point before applying blind identication algorithms to each segment separately. To the best of our knowledge, CPD has not been previously investigated in the context of channel coding, and this thesis presents the rst systematic study of this problem.

The primary contribution is a two-stage framework for estimating the change-point when the transmitter switches from one binary linear block code to another. In Stage 1, we apply a data processing strategy based on inner-product projections with vectors from the dual code, which transforms the received high-dimensional codeword sequence into a one-dimensional sequence of independent Bernoulli random variables while preserving the change-point signature. We then compute the maximum likelihood estimate of the change-point on this projected sequence and analytically construct a condence interval around the estimate such that the true change-point lies within the interval with a desired probability, using the asymptotic distribution of the estimator derived by Hinkley. In Stage 2, we restrict attention to the unprocessed received data within this condence interval and employ a trained one-dimensional convolutional neural network to rene the change-point estimate. The key insight is that the coarse statistical estimation narrows the search region, while the neural network provides ne-grained localization within this region. The combined two-stage pipeline signicantly outperforms either stage in isolation: the coarse maximum likelihood estimate alone achieves a detection probability of 0.54 at low noise levels, while the combined pipeline achieves 0.81 under the same conditions. We also address the practical scenario where the channel noise parameter is unknown, demonstrating that the framework maintains consistent detection accuracy across varying channel conditions without requiring prior knowledge of the noise level. The validity and performance of the proposed method are demonstrated through extensive simulations under both the binary symmetric channel and additive white Gaussian noise channel.

We also investigate two additional CPD-related problems. First, we study the problem of detecting when the transmitter switches between modulation schemes. We develop two online sliding-window detection methods: a feature-based method using higher-order cumulants as a discriminating statistic, and a likelihood-based method adapting the quasi Average Likelihood Ratio Test to a sliding-window framework. Both methods employ a conrmation mechanism to control the tradeo between detection delay and false alarm rate. This investigation is ongoing, and simulation results will be reported in a forthcoming publication. 

Finally, we demonstrate that the project then detect paradigm developed for communication signals generalizes to a fundamentally dierent domain: detecting authorship changes in multi-author documents. In this cross-domain application, we replace the dual-code projection with contrastive stylistic embeddings obtained from a fine-tuned transformer-based language model. Each paragraph of a document is mapped to a dense embedding vector, and the dissimilarity between consecutive paragraph embeddings produces a one-dimensional signal analogous to the projected Bernoulli sequence in the communication setting. Style change-points appear as spikes in this dissimilarity sequence, enabling the application of CPD methods. Preliminary results using a pairwise classier achieve F1 scores of 0.995, 0.92, and 0.81 on easy, medium, and hard benchmark datasets respectively, competitive with the state of the art. This cross-domain extension establishes the two-stage paradigm as a general-purpose framework for change-point detection in sequential data, regardless of the observation modality.

August 2026