Department of Electrical and Electronics Engineering
Permanent URI for this collectionhttp://localhost:4000/handle/123456789/1925
Browse
21 results
Search Results
Item A multi-modal smart switching based image transmission using semantic communication(IEEE, 2025-02) Tripathi, Sharda; Joshi, SandeepThe conventional paradigm of communication primarily concentrates on the transmission of raw data, often disregarding its contextual meaning. However, to tackle the exponential growth in data demands along with the limited availability of transmission bandwidth, there is an increasing need to transition from Shannon’s classical information-theoretic communication to a more advanced framework centered on semantics. This work presents a multi-modal semantic-based communication method for the transmission of high-definition images aimed at optimizing the transmitted data volume while maintaining a high throughput and mean intersection over union score. To this end, two architectural models are explored: a denser ResNet-based and a lightweight U-Net-based. Depending on the required QoS and resource availability, the raw image is either semantically segmented to obtain a fine-grained, pixel-level classification of the image or represented as label semantics, which provides only a higher-level, object-based, or region-based classification prior to its transmission. The experimental results show that such an adaptive semantic image processing approach leads to around 63% reduction in the transmitted data volume without compromising on the quality of image reconstruction.Item A quality-of-service-centric uplink rate-splitting approach for next-generation multiple access(IEEE, 2025-06) Tripathi, ShardaRecently, Rate-Splitting Multiple Access (RSMA) has emerged as a powerful paradigm for meeting the demanding performance requirements of 6G wireless networks through non-orthogonal high-rate data transmission. However, uplink access in RSMA necessitates optimizing the decoding order, which can lead to significant search latency. Besides, the process overlooks the Quality-of-Service (QoS) constraints of different traffic types, making current RSMA methods inadequate, especially for low-latency communication. Here, we address this issue by proposing QORA, short for QoS-aware One-shot Rate-splitting multiple Access, a multi-agent Deep Q-Network (DQN) framework that leverages a novel QoS-aware transmit power allocation and decoding order policy in uplink RSMA that achieves remarkable performance improvements while maintaining low latency and high admission rates.Item Sustainable marine surveillance sensor network aided by swipt-enabled auvs(IEEE, 2025-08) Tripathi, ShardaOceans are vital for Earth's climate stability, oxygen production, and as sources of food and energy for countless organisms. However, human-induced climate change significantly disrupts marine ecosystems, emphasizing the need for advanced underwater monitoring. The Internet of Underwater Things (IoUT), composed of marine sensor networks, offers a promising solution, nonetheless, challenges such as limited communication range and constrained power supplies. To address these issues, this work proposes using simultaneous wireless information and power transfer (SWIPT) from autonomous underwater vehicles (AUVs) to enhance sensor node efficiency. We formulate an integer linear programming (ILP) problem aimed at optimizing AUV trajectories through marine sensor networks, minimizing propulsion energy and mission duration while ensuring adequate energy harvesting at each node. The problem is proven to be NP-hard, resembling the well-known traveling salesman problem (TSP). Further, we introduce CLEAR, a multi-agent deep-Q-network (DQN) framework that effectively selects optimal path for AUV-based data muling. Experimental results demonstrate that CLEAR significantly improves network energy-efficiency, reduces mission duration, boosts harvested energy, and decreases age-of-information (AoI) of sensor data.Item When to Reach for the Skies? A DRL-Based Routing Framework for Non-Terrestrial Networks(IEEE, 2025-01) Tripathi, Sharda; Joshi, SandeepNon-terrestrial networks are envisioned to be an integral component of the beyond-fifth-generation wireless communication networks, catering to both conventional and emerging communication applications. In particular, a plethora of use cases are emerging for ultra-reliable low-latency communication, which require dynamic and quality of service compliant frameworks. In this letter, we formulate a binary integer non-linear programming problem to route time-critical traffic through non-terrestrial nodes. As the problem is NP-hard, we propose the solution using a deep reinforcement learning framework, taking into account the interactions between the terrestrial and various non-terrestrial nodes with an end-to-end latency target while maximizing the coverage probability. We perform simulations for multiple latency deadlines and outage thresholds and the results corroborate the efficiency of the proposed framework. Furthermore, we benchmark the proposed framework and show an improvement of 96.31% in coverage while incurring only 3.2% latency violations compared to the state-of-the-art.Item Data-driven Secure Authentication for Smart Grid IoT Networks(IEEE, 2023-07) Tripathi, ShardaInternet of Things (IoT) has found wide-spread usage in cyber physical systems. We consider the security aspects related to cyber physical systems, particularly the smart grid, wherein IoT devices are deployed for monitoring and control, often in the open without any active surveillance, thereby increasing their vulnerability to security attacks. To this end, device authentication is an essential security feature for smart grid IoT networks. A desirable authentication protocol designed for IoT devices should not only be robust when the device is compromised by an adversary, it also needs to be computationally efficient in the wake of limited storage and computational capabilities of IoT devices. In this work, we address the critical issue of device security in smart grid IoT networks, and propose a data-driven, yet lightweight and privacy preserving authentication scheme for IoT devices. The performance analysis based on our experiments show that the proposed scheme is robust against heterogeneous network attacks, and significantly reduces computational load.Item MERGE: Meta Reinforcement Learning for Tunable RL Agents at the Edge(IEEE, 2023-12) Tripathi, ShardaThe efficient allocation of radio resources is an essential trait of 5G/6G radio access networks (RANs), as they are called to meet diverse QoS requirements of highly demanding applications. To equip RANs with such an ability and, at the same time, meet their function split constraints, we envision a distributed learning approach for radio resource allocation that makes the most out of the Central Unit (CU) and Distributed Unit (DU) components by effectively exploiting their synergy. On the one hand, our solution, named MERGE, leverages the knowledge of the radio connectivity dynamics that each DU can acquire through the local use of a deep reinforcement learning radio agent. On the other hand, it lets the CU collect such agents in a crowdsourcing fashion, and, then, thanks to a meta-learning policy, properly select and aggregate them to create up-to-date radio agents of the right size (hence, complexity level) to fit the computing constraints of the individual DUs. Our results show that MERGE can match the performance of the highest-complexity radio model in [1] with 25% less computational requirements, and, for a given computational resource, it outperforms a single pruned model with a 19% increase in QoS.Item EnRoute: A DQN based Energy Efficient Routing for URLLC in Next Generation Networks(IEEE, 2023) Tripathi, Sharda5G networks and beyond (B5G) will support a wide range of services which demand large data rates, massive connection densities, high reliability and low latency. Catering to such needs will require a dense deployment of gNodeBs (gNBs), leading to enormous network energy consumption, thereby contributing to the global carbon emissions. Conventional network energy saving techniques, such as gNB sleep scheduling, significantly hamper the latency performance of delay-sensitive data. To this end, we propose EnRoute for energy efficient route selection of delay-sensitive traffic in B5G networks. We formulate an integer programming problem to optimally select a routing path for delay sensitive data such that its target delay is met while maximizing the energy efficiency of gNBs. Since the problem is NP-complete, we resort to DQN-based learning framework for designing EnRoute. Our simulation results confirm an energy efficient route selection and the key performance indices meeting the required QoS targets. We further compare our approach with two benchmark schemes, one that minimizes the latency, and other that maximizes the energy efficiency of gNBs (instead of tackling them together) for the route selection, and show that our latency performance matches the former (99.7%), with a marginal deficit in energy efficiency with respect to latter (10%).Item Versatile Multivariate Data Pruning in Smart Grid IoT Networks(IEEE, 2020) Tripathi, ShardaWith wide scale sensor deployments in smart grid IoT networks, there has been a manyfold increase in the variety and quantity of data generated in the network. In this work, the problem of data reduction in smart grid IoT network is addressed to enhance the resource utilization without hampering the required quality of service. A novel versatile algorithm for multivariate data pruning at the edge devices in smart grid IoT networks is presented. This is achieved via a two stage data reduction mechanism which first exploits the inter-variable correlation to cut down on the number of transmitted variables, followed by adaptive data compression in temporal domain using adaptive compressive sampling. It is shown that with the application of the proposed algorithm at the edge nodes, around 23% savings in bandwidth requirement can be achieved with minimum loss of information.Item Pathway and Future of IoE in Smart Cities: Challenges of Big Data and Energy Sustainability(CRC, 2021) Tripathi, ShardaThis chapter deals with the IoT node-level and network energy sustainability and IoE stability aspects, wherein the benefits from node-level and network intelligence are presented. Some field IoT sensing applications, such as in smart cities, pollution monitoring, and agricultural automation, are characterized by high energy consumption, node lifetime criticality, and energy sustainability issues. More discussion on smart meter data compressibility and novel approaches to dealing with big data concerns caused by millions of smart meter installation as a part of smart city monitoring exercise are presented in the chapter. It has been demonstrated how the challenges of big data and energy sustainability are addressed via application-specific unique approaches. The research state-of-the-art and open issues on mitigating big data challenges and energy sustainability have been highlighted. Further, the role of smart grid on uninterrupted IoE operation, the benefits of distributed energy generation through IoE, and the newer challenges of power grid stability/controllability have also been discussed.Item Fair and Scalable Orchestration of Network and Compute Resources for Virtual Edge Services(IEEE, 2023-03) Tripathi, ShardaThe combination of service virtualization and edge computing allows for low latency services, while keeping data storage and processing local. However, given the limited resources available at the edge, a conflict in resource usage arises when both virtualized user applications and network functions need to be supported. Further, the concurrent resource request by user applications and network functions is often entangled, since the data generated by the former has to be transferred by the latter, and vice versa. In this paper, we first show through experimental tests the correlation between a video-based application and a vRAN. Then, owing to the complex involved dynamics, we develop a scalable reinforcement learning framework for resource orchestration at the edge, which leverages a Pareto analysis for provable fair and efficient decisions. We validate our framework, named VERA, through a real-time proof-of-concept implementation, which we also use to obtain datasets reporting real-world operational conditions and performance. Using such experimental datasets, we demonstrate that VERA meets the KPI targets for over 96% of the observation period and performs similarly when executed in our real-time implementation, with KPI differences below 12.4%. Further, its scaling cost is 54% lower than a centralized framework based on deep-Q networks
- «
- 1 (current)
- 2
- 3
- »