Shannon Wireless’ Post

The #channel #state #information (#CSI) measured by a #WiFi receiver suffers from errors in both its gain and phase, which can significantly hinder sensing tasks. By analyzing these errors from different WiFi receivers, a mathematical model for these gain and phase errors is developed in this work. Based on these models, several theoretically justified #preprocessing #algorithms for correcting such errors at a receiver and, thus, obtaining clean CSI are presented. ---- Vishnu Vardhan Ratnam, Hao Chen, Hao-Hsuan Chang, Abhishek Sehgal, Charlie Zhang More details can be found at this link: https://lnkd.in/eRC-_z95

🌟 Publication Alert 🌟 I’m thrilled to share my first IEEE magzine in the prestigious 𝐈𝐄𝐄𝐄 𝐂𝐨𝐧𝐬𝐮𝐦𝐞𝐫 𝐄𝐥𝐞𝐜𝐭𝐫𝐨𝐧𝐢𝐜𝐬 𝐌𝐚𝐠𝐚𝐳𝐢𝐧𝐞. According to 𝑮𝒐𝒐𝒈𝒍𝒆 𝑺𝒄𝒉𝒐𝒍𝒂𝒓, it is 𝒏𝒖𝒎𝒃𝒆𝒓 𝒐𝒏𝒆 venue for consumer electronics (https://lnkd.in/dmhF8U_x) 📄 𝐓𝐢𝐭𝐥𝐞: Enhancing Consumer Privacy in Federated Load Forecasting Through Single Layer Aggregation 🔍 𝐓𝐡𝐞 𝐏𝐫𝐨𝐛𝐥𝐞𝐦: While differential privacy enhances the privacy of AI models, it often comes at the cost of reduced accuracy particularly in federated load forecasting for energy networks. 💡 𝐎𝐮𝐫 𝐒𝐨𝐥𝐮𝐭𝐢𝐨𝐧: We proposed a novel single-layer aggregation framework that minimizes the negative impact of differential privacy on accuracy while maintaining robust privacy protection. This innovative approach strikes a balance between privacy and performance, opening doors to more efficient and secure federated learning applications. 🙏 𝐀𝐜𝐤𝐧𝐨𝐰𝐥𝐞𝐝𝐠𝐦𝐞𝐧𝐭𝐬: This work would not have been possible without the invaluable support and collaboration of my co-authors Dr. Sanaullah Manzoor, Dr. Muhammad Ali Jamshed, and my supervisor Dr. Ahmed Zoha. Thank you for your guidance and encouragement throughout this journey! 𝐥𝐢𝐧𝐤:  https://lnkd.in/dvxhe_Yz https://lnkd.in/dn4QNCQb #AI #FederatedLearning #DifferentialPrivacy #ConsumerElectronics #LoadForecasting #IEEE

Over the past few years, our focus has been on addressing key questions related to enhancing microgrid resiliency through algorithmic innovations in mitigation strategies and bridging the sim-to-real gap in deploying learning-based policies. We have developed hierarchical resilient control layers using a new vertical variant of federated RL to tackle issues regarding security and privacy under multi-party microgrid ownership and with limited knowledge about proprietary inverter controls. Once the policies are learned offline using simulators (e.g., GridLAB-D), subsequent efforts are geared toward transferring these to the real-time testbed developed using high-fidelity platforms such as Hypersim that can achieve desirable performance and thus provide operators more confidence in deploying neural policies. These aspects are described in detail in our recent IEEE Trans. on Smart Grid article (early access: https://lnkd.in/g7mjUWRa). We greatly acknowledge the support provided by PNNL's Resilience through Data-Driven, Intelligently Designed Control (RD2C) initiative. Wonderful effort by the PNNL team and collaborators Ramij Raja Hossain, Sheik Mohammad Mohiuddin, Yuan Liu, Wei Du (project PI), Veronica Adetola (RD2C S&T thrust lead), Rohit Jinsiwale, Qiuhua Huang, Tianzhixi (Tim) Yin, and Ankit Singhal.

Interesting article that provides a short LLMs overview and indicates how they offer transformative potential in telecom by supporting anomaly resolution, 3GPP standards comprehension, and network optimization. LLMs excel in semantic understanding, knowledge retrieval, and task orchestration but face challenges like hallucinations, lack of explainability, and computational demands. It highlights the need for telecom-specific models, privacy safeguards, and compression techniques for effective deployment. Real-time updates and sustainable designs are crucial for enhancing their efficiency and relevance. LLMs can significantly improve operational efficiency and decision-making in the telecom industry.

Read the first page of “AI-RAN in 6G Networks: State-of-the-Art and Challenges” for free! This paper provides a detailed survey and thorough assessment of AI-RAN’s vision and state-of-the-art challenges. Read now: https://bit.ly/3Z6TVGG

🌟 Excited to announce the publication of our latest research paper titled "Advancements in Overlapping Sound Event Detection" in the proceedings of the 14th International Conference on Cloud Computing, Data Science & Engineering (Confluence). 📝 Our team delved into the complexities of simultaneous sound events and explored innovative solutions to enhance detection accuracy and robustness. 🔍 In this paper, we conducted a comprehensive review of existing methodologies, dissecting their strengths and limitations, and addressing challenges in evaluation metrics and datasets. Our research contributes to the advancement of sound event detection technology, particularly in complex audio environments, paving the way for better solutions in real-world applications. 📅 Conference Date: 18-19 January 2024 📌 Published in: IEEE Xplore 🔗 DOI: 10.1109/Confluence60223.2024.10463375 👉 Read the full paper for insights into the evolving landscape of sound event detection and stay tuned for more updates on our research journey! #Research #SoundEventDetection #Innovation #ConferencePublication #IEEE #TechInnovation #AI #MachineLearning #DataScience 🚀🔊🎶

AI-optimized and distributed cognitive radio networks will be fundamental for connecting the other half (including those that appear in the statistics as internet users but with a big gap in terms of QoS). https://lnkd.in/gGEkJ5aU #ConnectingTheUnconnected IEEE

In this paper, they systematically review the latest developments in quantum Internet protocols from the perspective of protocol stack layering.With the development of #quantum #technologies, the #quantum #Internet has demonstrated unique applications beyond the classical Internet and has been investigated extensively in recent years. In the construction of #conventional #Internet #software, the protocol stack is the core architecture for #coordinating #modules. How to design a protocol stack for the #quantum #Internet is a challenging problem. By summarizing and analyzing the progress in each layer’s protocols, they reveal the current research status and connections among the layers. ----@Yuan Li; @Hao Zhang; @Chen Zhang; @Tao Huang; F. Richard Yu More details can be found at this link: https://lnkd.in/g_fxbXTr

🎉 Exciting News! 🎉 I'm thrilled to announce that our paper, titled "A DDPG-based Zero-Touch Dynamic Prioritization to Address Starvation of Services for Deploying Microservices-based VNFs," has been successfully accepted for publication in the IEEE Transactions on Machine Learning in Communications and Networking. This paper introduces innovative solutions to enhance resource allocation efficiency in the realm of Network Function Virtualization (NFV), addressing critical problems such as dynamic service prioritization and the notorious low-priority service starvation issue. By leveraging a Deep Deterministic Policy Gradient (DDPG) model, we've developed a zero-touch framework that dynamically prioritizes services to optimize network function deployment. A big thank you to my co-authors Avishek Nag, PhD, Hamed Ahmadi for the support and contribution to this project. We are looking forward to seeing the impact this will have on the industry and continuing our work to innovate and improve network technologies. https://lnkd.in/exfcNG_9 Stay tuned for more updates! #IEEE #MachineLearning #Telecommunications #5G #NetworkFunctionVirtualization #Research #Innovation

This paper formulates a novel methodology for #anticipatorily #allocating #communication and computational resources at the network edge, based on the prediction of spatio-temporal dynamics of mobile users. The conceived architecture exploits a Software-Defined Networking approach to monitor users’ mobility, a Convolutional #Long #Short-#Term #Memory to predict over different look-ahead horizons the number of users within a given number of cells and their related service demands, and Dynamic Programming to optimally allocate users’ requests among available #Multi-#access #Edge #Computing servers. Computer simulations investigate the effectiveness of the proposed approach in a realistic autonomous driving use case and compare its behavior against a baseline solution. ---- Arcangela Rago, Giuseppe Piro, Gennaro Boggia, Paolo Dini More details can be found at this link: https://lnkd.in/ey-SzPhw