#energy efficiency
25 resultsA Stochastic Optimization Framework for RIS-Aided Wireless Network Design
Davide Gagliardi, Alessio Zappone, Domenico Ciuonzo +1
The paper proposes a stochastic optimization framework using continuous cross‑entropy and Metropolis‑Hastings methods to design reconfigurable intelligent surface (RIS) configurati…
LightRot: A Light-Weighted Rotation Scheme and Architecture for Accurate Low-Bit Large Language Model Inference
Sangjin Kim, Yuseon Choi, Jungjun Oh +2
LightRot introduces a lightweight rotation scheme and a dedicated hardware accelerator that enable energy‑efficient, low‑bit inference for large language models such as LLaMA2‑13B…
Energy-Efficient Access-Point Sleep-Mode Techniques for Cell-Free mmWave Massive MIMO Networks With Non-Uniform Spatial Traffic Density
Jan GarcÃa-Morales, Guillem Femenias, Felip Riera-Palou
The paper proposes energy‑efficient sleep‑mode strategies for access points in cell‑free mmWave massive MIMO networks that account for realistic, non‑uniform spatial traffic distri…
The Sparsity Ceiling: Where Spiking Networks Can and Cannot Trade Activity for Energy
Zeyu Wang
The paper studies how much spiking neural networks can lower their firing activity without losing performance, showing that the achievable sparsity depends on the task and architec…
The Price of Meaning: Quantifying Semantic Communication Overheads in Practice
Xinyi Lin, Peizheng Li, Adnan Aijaz
The paper presents an analytical framework to quantify the spectral and energy overheads of semantic communication, identifying conditions under which semantic compression yields n…
LLMET: Enabling Cross-Layer Evaluation of Emerging M3D Memories for Energy-Efficient LLM Serving
Ming-Yen Lee, Hanchen Yang, Faaiq Waqar +4
The paper introduces LLMET, a cross‑layer simulation framework that evaluates how emerging monolithic 3D (M3D) on‑chip memory can cut energy use when serving large language models,…
Secure Energy-Efficient Uplink Transmission in Movable-Element RIS-aided Systems with Movable Antennas and Artificial Noise
Ayda Nodel Hokmabadi, Mohamed Elhattab, Chadi Assi
The paper proposes a method to maximize secure energy efficiency in uplink transmissions by jointly optimizing transmit powers, receive processing, artificial noise, RIS phase shif…
Evolutionary AP Switch ON/OFF Techniques for Energy-efficient Cell-free Massive MIMO Networks
Jan GarcÃa-Morales, Alejandro de la Fuente, David Gualda +3
The paper proposes two evolutionary algorithms—a constrained genetic algorithm and a Pareto-driven genetic algorithm—to select active access points in cell‑free massive MIMO networ…
Adaptive Sampling for Spatiotemporal Anomaly Monitoring in Wireless Sensor Networks
Guoqing Lu, Yixuan Sun, Yiwen Jiang +1
The paper introduces a sentinel‑assisted adaptive sampling framework for wireless sensor networks that combines Kalman‑filter‑driven sparse sensing with sentinel nodes performing G…
Valinor: Architectural Support for Fast, Energy-Efficient and Programmable Physical Memory Allocation
Konstantinos Kanellopoulos, Spiros Galanopoulos, Konstantinos Sgouras +7
Valinor is a hardware‑OS cooperative substrate that provides a programmable allocation engine to accelerate physical memory allocation, achieving hardware‑level speed while retaini…
Energy-Efficient Federated Learning via Adaptive Encoder Freezing for MRI-to-CT Conversion: A Green AI-Guided Research
Ciro Benito Raggio, Lucia Migliorelli, Nils Skupien +6
The paper proposes an adaptive encoder‑freezing technique for federated learning that reduces energy use and CO2 emissions while preserving MRI‑to‑CT conversion quality.
Latency Coding for Efficient and Low-Latency Deep Spiking Neural Networks
Yi Lu, Jianhao Ding, Zhaofei Yu
The paper introduces latency coding, an extension of time‑to‑first‑spike coding, and a training framework using backpropagation through time to build deep spiking neural networks t…
Architectural Trade-offs in the Energy-Efficient Era: A Comparative Study of power-capping NVIDIA H100 and H200
Aditya Ujeniya, Jan Eitzinger, Georg Hager +1
The paper compares NVIDIA H100 and H200 GPUs under different power caps, focusing on how memory bandwidth and power distribution affect performance per watt for compute‑bound and m…
GNN-based Online Beamforming Design for HAPS-Assisted NTN
Lavanya S S Anjapuli, Animesh Yadav, Halim Yanikomeroglu
The paper proposes using a high‑altitude platform station (HAPS) to relay data for cell‑edge users and designs beamforming vectors at both the terrestrial base station and HAPS via…
CIMERA: Compute-in-Interconnect and Memory with Reconfigurable Precision for LLM Inference
Yue Jiet Chong, Yimin Wang, Wei Zhang +1
The paper introduces CIMERA, a hardware accelerator that integrates compute-in-interconnect and memory with reconfigurable precision to run large language models more energy‑effici…
Goal-Oriented Sensor Reporting Scheduling for Non-linear Dynamic System Monitoring
Prasoon Raghuwanshi, Onel Luis Alcaraz López, I-Hong Hou +2
The paper proposes a goal‑oriented scheduling framework for IoT sensors monitoring a non‑linear dynamic system, using deep reinforcement learning and LSTM‑based query prediction to…
Adaptive Federated Learning to Optimize Integrated Flows in Cyber-Physical Data Centers
Junhong Liu, Lanxin Du, Yujia Li +4
The paper proposes an adaptive federated learning framework, combined with cryptographic techniques, to efficiently and privately optimize electricity, heat, and data flows across…
Green Deep Reinforcement Learning for IoT Edge Intrusion Detection
Saeid Jamshidi, Foutse Khomh, Rolando Herrero +2
The paper introduces two carbon‑aware deep reinforcement learning intrusion detection systems for IoT edge gateways, one label‑free (DeepEdgeIDS) and one supervised (AutoDRL‑IDS),…
Schwarz Information Criterion Aided MAB for Resource Allocation in Dynamic LoRa System
Ryotai Ariyoshi, Aohan Li, Mikio Hasegawa +3
The paper introduces a lightweight distributed learning approach that combines UCB1‑tuned multi‑armed bandit selection with the Schwarz Information Criterion to quickly adapt LoRa…
MelT: A Portable, Single-GEMM Mel Audio Frontend via Non-Uniform DFT with Measured Latency and Energy Gains on GPUs
Augusto Camargo, Marcelo Finger
The paper introduces MelT, a single‑stage Mel audio frontend that replaces the traditional STFT‑based pipeline with a precomputed non‑uniform DFT applied via GEMM, achieving 1.6‑3.…
Learning-Based Beamforming for Energy Efficiency of Continuous Aperture Array Systems
Shiyong Chen, Jia Guo, Shengqian Han
The paper proposes a learning framework that jointly determines the size of continuous aperture arrays and their beamforming functions to maximize downlink energy efficiency, using…
EMO: Energy Efficiency Modeling and Optimization for AI Workloads
Jiyu Luo, Shaoyu Chen, Jingwei Sun +3
EMO is a lightweight framework that models and optimizes the energy consumption of GPU-accelerated AI workloads by detecting fine‑grained slack in asynchronous execution and applyi…
A Comprehensive Evaluation of Deep Learning Object Detection Models on Heterogeneous Edge Devices
Daghash K. Alqahtani, Muhammad Aamir Cheema, Maria A. Rodriguez +1
The paper benchmarks several deep learning object detection models on various edge devices, measuring accuracy, latency, and energy use while also analyzing performance as scene co…
Enabling Energy-Efficient Simultaneous Multi-Task Reinforcement Learning through Spiking Neural Networks with Active Dendrites for Bio-inspired Generalist Agents
Rachmad Vidya Wicaksana Putra, Avaneesh Devkota, Muhammad Shafique
The paper introduces MTSpark, a method that combines spiking neural networks with active dendrites to enable energy‑efficient simultaneous multi‑task reinforcement learning, achiev…