Reshma Prasad

Journals & Magazines

M. Elkael, M. Polese, R. Prasad, S. Maxenti, and T. Melodia. “ALLSTaR — Automated LLM-Driven Scheduler Generation and Testing for Intent-Based RAN.” IEEE Transactions on Mobile Computing (2026)Journal

DOI Link

The evolution toward open, programmable O-RAN and AI-RAN 6G networks creates unprecedented opportunities for Intent-Based Networking (IBN) to dynamically optimize Radio Access Network (RAN) operations based on dynamic operators requirements. However, applying IBN effectively to the RAN scheduler - a critical component determining resource allocation and system performance - remains a significant challenge. Current approaches predominantly rely on coarse-grained network slicing, lacking the granularity for dynamic adaptation to individual user conditions and traffic patterns. Despite the existence of a vast body of scheduling algorithms that could potentially translate high-level intents into executable policies, their practical utilization is hindered by implementation heterogeneity, insufficient systematic evaluation in production environments, and the complexity of developing high-performance scheduler implementations. This necessitates a more granular, flexible, and verifiable approach to align scheduler behavior with operator-defined intents. To address these limitations, we propose ALLSTaR (Automated LLm-driven Scheduler generation and Testing for intent-based RAN), a novel framework leveraging Large Language Models (LLMs) for automated, intent-driven scheduler design, implementation, and evaluation. ALLSTaR interprets natural language intents, automatically generates functional scheduler code from the research literature using Optical Character Recognition (OCR) and LLMs, and intelligently matches operator intents to the most suitable scheduler(s). Our implementation deploys these schedulers as O-RAN dApps, enabling on-the-fly deployment and comprehensive testing on a production-grade, multi-vendor 5G-compliant testbed. This approach has enabled the most extensive Over-The-Air (OTA) experimental comparison of scheduling algorithms to date, evaluating 18 distinct algorithms synthesized from the academic literature and validated through the testing pipeline. The resulting performance profiles serve as the input for our Intent-Based Scheduling (IBS) framework, which dynamically selects and deploys appropriate schedulers that optimally satisfy operator intents.We validate our approach through multiple use cases unattainable with current slicing-based optimization techniques, demonstrating fine-grained control based on buffer status, physical layer conditions, and heterogeneous traffic types.

Conference Papers

R. Shirkhani, R. Prasad, L. Bonati, T. Melodia, and M. Polese. “RANalyzer: Automated Continuous RAN Software Evaluation and Regression Analysis.” 2026 IEEE 12th International Conference on Network Softwarization (NetSoft) (2026)Conference

DOI Link

Software-driven O-RAN architectures enable rapid innovation through frequent, independent updates to virtualized components. However, attributing performance variations to specific software changes is challenging due to the stochastic nature of wireless systems, where channel conditions, interference, and hardware variability confound analysis. Traditional threshold-based monitoring and manual troubleshooting do not scale with modern software evolution. This paper presents RANalyzer, an automated test analysis framework that quantifies the performance impact of software updates beyond what can be explained by wireless channel conditions. RANalyzer combines LLM-assisted semantic extraction with residuals analysis. The first categorizes code changes by affected protocol layers and functional components, while the second provides insights on the effect of load, channel, or code changes on the test performance. We contribute an extensive dataset collected over more than two years of continuous over-the-air testing on an experimental O-RAN testbed, comprising over 8,600 automated tests across 69 releases of the OpenAirInterface (OAI) stack. By modeling expected performance and interpreting deviations as software-induced effects, we identify degraded instances attributable to code changes and correlate them with specific change categories. The framework can be integrated into CI/CD/CT pipelines for automated, continuous evaluation of software updates at scale.

Journals & Magazines

Journals & Magazines

Journals & Magazines

R. Prasad and A. Sunny. “Scheduling Slice Requests in 5G Networks.” IEEE/ACM Transactions on Networking (2023)Journal

DOI Link

Network slicing is a 5G paradigm that enables the creation of on-demand logical networks over shared physical infrastructure. In this paper, we present a framework that allows users to make advance slice reservations with the End-to-End Orchestrator (EEO). Our reservation mechanism enables the EEO to make admission decisions instantly upon request arrival, providing guarantees as to when the request can be enabled. We then proceed to address a relevant revenue maximization problem through an optimal solution, which has factorial time complexity. We also propose a low-complexity algorithm that can efficiently allocate resources for the online version of the problem. We conduct evaluations that demonstrate how the reservation mechanism can potentially improve EEO’s revenue. Additionally, we conduct a study on scenarios where the arrival rates of slice requests exhibit a positive correlation with reservation discounts provided by EEO.