Computational Energy Consumption Research Articles

Energy efficiency analysis of spin–orbit torque MRAM using composition dependent resistivity and spin Hall angle in Pt/W multilayer structure

Current-induced magnetization switching through spin–orbit torque (SOT) enables low-energy and high-speed switching of magnetic memory (MRAM), positioning SOT-MRAM as a promising candidate for next-generation memory technology. As energy consumption in data computation has become a major challenge in the 21st century, significant efforts have been dedicated to developing energy-efficient memory devices. To optimize SOT-MRAM for energy efficiency, achieving a high spin Hall angle (SHA) has been regarded as essential, as a high SHA indicates efficient spin current generation. However, while previous studies have focused on increasing SHA, this has often led to increased resistivity, which may hinder overall energy efficiency in SOT-MRAM. We propose that resistivity is almost as important as SHA in determining the energy efficiency of SOT-MRAM. In this study, we analyze the effects of both SHA and resistivity in SOT channel materials, which can be modulated by adjusting the composition of Pt and W in Pt/W multilayers. We observed that varying the Pt and W compositions resulted in non-linear changes in both SHA and resistivity, suggesting that evaluating energy efficiency based solely on SHA may be misleading. Thus, we introduce a figure of merit, ξ=ρθSH2, which represents the combined impact of SHA and resistivity on SOT-MRAM energy efficiency. We hope that our findings reveal the importance of considering both SHA and resistivity in the development of energy-efficient SOT-MRAM.

Real-time discrimination of earthquake signals by integrating artificial intelligence technology into IoT devices

The real-time detection and analysis of seismic signals is crucial in geophysics research, especially when it comes to monitoring catastrophic events. We present an evolutionary deep learning method that yields a model named MCU-Quake. This model encodes the discrimination process as a single numerical value, offering interpretability with only 2693 parameters. Trained on raw seismic waveforms from Utah, USA, MCU-Quake demonstrates its generalization capability across a global natural earthquake dataset. Notably, the model effectively identifies typical explosions during the Russia-Ukraine war in Europe. The knowledge to discriminate between ambient noise, explosions and natural earthquakes can be represented by values of −5.01 (std: 1.14), 1.96 (std: 0.36), 1.01 (std: 0.49), respectively. The model can be deployed on Internet of Things (IoT) devices, including most microcontrollers, which are constrained by limited computational resources (kilo-bytes of memory) and energy consumption (micro-Watts). The results indicate the prospect of on-site missions of artificial intelligent sensors.

TinySpiking: a lightweight and efficient python framework for unsupervised learning spiking neural networks

Abstract Neural computation frameworks are essential for advancing computational neuroscience and artificial intelligence, offering a robust platform for simulating intricate brain-like processes and fostering the growth of intelligent systems. This study presents TinySpiking, a novel, lightweight, and energy-efficient Python framework designed for the simulation and learning of Spiking Neural Networks (SNNs). Unlike traditional frameworks, TinySpiking does not depend on third-party libraries, thereby reducing computational overhead and energy consumption. The framework's innovation is that it implements unsupervised learning through Spike-Time Dependent Plasticity (STDP), a biologically inspired learning rule that adjusts synaptic weights based on the precise timing of neuronal spikes. This mechanism is crucial for constructing complex neural architectures and effectively processing spatio-temporal data, which is vital for the dynamic and intricate demands of real-world data analysis. Demonstrating the practical utility of TinySpiking, we applied it to image reconstruction tasks using the MNIST and landmark datasets. The results not only validate the network's ability to autonomously identify and enhance key visual features without external supervision but also highlight its efficiency in learning from data, mirroring the adaptability of biological neural systems. In conclusion, TinySpiking's innovative, lightweight design and its proven effectiveness in unsupervised learning tasks make it a standout computational tool for the fields of computational neuroscience and machine learning. Its low time consumption, biological plausibility, and independence from third-party libraries position it as a compelling platform for future research and applications, promising to drive advancements in unsupervised learning and intelligent system development.

Efficient Locomotion for Space Robots Inspired by the Flying Snake

Robots are becoming an integral part of space facilities construction and maintenance, and may need to move to and from different work locations. Robotic arms that are widely employed, which are mounted on fixed bases, have difficulty coping with increasingly complex missions. The challenge discussed in this paper is the problem of the efficient locomotion of robotic systems. Inspired by the gliding motion of a flying snake launched from a tree and combined with the weightlessness of the space environment, we design similar motions for the robot, including the following three steps. First, the robot folds its body like a snake and initiates flight by accelerating the global center of mass (CM), focusing on the movement direction and generating suitable momentum. Then, during the flight (free-floating) phase, the joint motions are planned using a nonlinear optimization technique, considering the nonholonomic constraints introduced by the momentum conservation and the system states at the initial and final states of the floating. Meanwhile, the difficulties caused by long-distance flights are addressed to reduce the motion computational cost and energy consumption by introducing the phase plane analysis method. Finally, the landing motion is designed to avoid rigid collisions and rollover on the radial plane. The numerical simulations illustrate that the three phases of maneuvers are smooth and continuous, which can help the space robots efficiently traverse the environment.

Synergizing Federated AI Systems with Circular Economy Principles: A Framework for Sustainable and Resilient Data Science

The convergence of Federated Artificial Intelligence (AI) systems and Circular Economy (CE) principles marks a transformative approach to sustainable and resilient data science. Federated AI, with its decentralized architecture and emphasis on data privacy, seamlessly integrates with CE's objectives of maximizing resource efficiency, reducing waste, and fostering closed-loop systems. This paper proposes a novel framework that synergizes CE principles with Federated AI to address persistent challenges, including data siloing, scalability, energy efficiency, and environmental sustainability. The framework leverages adaptive learning models and resource-aware algorithms to optimize data- driven decision-making across applications such as smart city development, sustainable supply chain management, and renewable energy optimization. Through rigorous simulations and real-world case studies, the study demonstrates measurable improvements: a 20–30% increase in resource efficiency and a marked reduction in computational energy consumption compared to traditional centralized AI systems. These findings highlight the transformative potential of Federated AI in driving circular and sustainable ecosystems. The research also contributes to ethical AI discourse, providing actionable insights for policymakers, academics, and industry leaders to harmonize AI advancements with global sustainability imperatives.

Design and Performance Analysis of a SPECK-Based Lightweight Hash Function

In recent years, hash algorithms have been used frequently in many areas, such as digital signature, blockchain, and IoT applications. Standard cryptographic hash functions, including traditional algorithms such as SHA-1 and MD5, are generally computationally intensive. A principal approach to improving the security and efficiency of hash algorithms is the integration of lightweight algorithms, which are designed to minimize computational overhead, into their architectural framework. This article proposes a new hash algorithm based on lightweight encryption. A new design for the lightweight hash function is proposed to improve its efficiency and meet security requirements. In particular, efficiency reduces computational load, energy consumption, and processing time for resource-constrained environments such as IoT devices. Security requirements focus on ensuring properties such as collision resistance, pre-image resistance, and distribution of modified bit numbers to ensure reliable performance while preserving the robustness of the algorithm. The proposed design incorporates the SPECK lightweight encryption algorithm to improve the structure of the algorithm, ensuring robust mixing and security through confusion and diffusion, while improving processing speed. Performance and efficiency tests were conducted to evaluate the proposed algorithm, and the results were compared with commonly used hash algorithms in the literature. The test results show that the new lightweight hash algorithm has successfully passed security tests, including collision resistance, pre-image resistance, sensitivity, and distribution of hash values, while outperforming other commonly used algorithms regarding execution time.

REHAS: Robust and Efficient Hyperelliptic Curve‐Based Authentication Scheme for Internet of Drones

ABSTRACTInternet of Drones (IoD) is one of the most beneficial and has many versatile applications like Surveillance and Security, Delivery and Logistics, Environmental Monitoring, Agriculture, and so forth. The IoD network is crucial for collecting sensitive data like geo‐coordinates, vehicle traffic data, and property details while surveying the various deployment locations in smart cities. The communication between users and drones can be compromised over insecure wireless channels by multiple attacks such as Man‐in‐the‐middle‐attack, Denial of Service, and so forth. Many schemes have already been propounded in the field of IoD. Still, many of them cannot address the resource constraints problem of drones, and existing protocols have higher computation and communication costs. Therefore, this paper has proposed a robust and efficient Hyper‐Elliptic Curve‐based authentication scheme (REHAS), which provides a session key for secure communication. Artificial Identities are generated using a hash function and random numbers. Fuzzy Extractor is used for user biometric authentication, which makes the smart device secure when lost. HECC is used with a smaller bit size of 80 bits rather than ECC of 160 bits. The security of the REHAS has been ensured using Scyther simulation. Furthermore, the resilience, safety, and robustness of REHAS are ensured by Informal security analysis. Lastly, a comparative study of the REHAS has been performed with other related Authentication and key agreement (AKA) protocols regarding communication cost, Computation cost, and security features, demonstrating that REHAS incurred less computation cost (6.7171 ms), communication overhead (1696 bits), and energy consumption (22.5 mJ) than other existing AKA schemes.

Resource-efficient artificial intelligence for battery capacity estimation using convolutional FlashAttention fusion networks

Accurate battery capacity estimation is crucial for optimizing lifespan and monitoring health conditions. Deep learning has made notable strides in addressing long-standing issues in the artificial intelligence community. However, large AI models often face challenges such as high computational resource consumption, extended training times, and elevated deployment costs. To address these issues, we developed an efficient end-to-end hybrid fusion neural network model. This model combines FlashAttention-2 with local feature extraction through convolutional neural networks (CNNs), significantly reducing memory usage and computational demands while maintaining precise and efficient health estimation. For practical implementation, the model uses only basic parameters, such as voltage and charge, and employs partial charging data (from 80 % SOC to the upper limit voltage) as features, without requiring complex feature engineering. We evaluated the model using three datasets: 77 lithium iron phosphate (LFP) cells, 16 nickel cobalt aluminum (NCA) cells, and 50 nickel cobalt manganese (NCM) oxide cells. For LFP battery health estimation, the model achieved a root mean square error of 0.109 %, a coefficient of determination of 0.99, and a mean absolute percentage error of 0.096 %. Moreover, the proposed convolutional and flash-attention fusion networks deliver an average inference time of 57 milliseconds for health diagnosis across the full battery life cycle (approximately 1898 cycles per cell). The resource-efficient AI (REAI) model operates at an average of 1.36 billion floating point operations per second (FLOPs), with GPU power consumption of 17W and memory usage of 403 MB. This significantly outperforms the Transformer model with vanilla attention. Furthermore, the multi-fusion model proved to be a powerful tool for evaluating capacity in NCA and NCM cells using transfer learning. The results emphasize its ability to reduce computational complexity, energy consumption, and memory usage, while maintaining high accuracy and robust generalization capabilities.

Scalable Microscale Artificial Synapses of Lead Halide Perovskite with Femtojoule Energy Consumption.

The efficient conduction of mobile ions in halide perovskites is highly promising for artificial synapses (or memristive devices), devices with a conductivity that can be varied by applying a bias voltage. Here we address the challenge of downscaling halide perovskite-based artificial synapses to achieve low energy consumption and allow high-density integration. We fabricate halide perovskite artificial synapses in a back-contacted architecture to achieve microscale devices despite the high solubility of halide perovskites in polar solvents that are commonly used in lithography. The energy consumption of a conductance change of the device is as low as 640 fJ, among the lowest reported for two-terminal halide perovskite artificial synapses so far. Moreover, the high resistance of the device up to hundreds of megaohms, low operating voltage of 100 mV and simple two-terminal architecture enable implementation in highly dense crossbar arrays. These arrays could potentially show orders of magnitude lower energy consumption for computation compared to conventional digital computers.

Analysing the radiation reliability, performance and energy consumption of low-power SoC through heterogeneous parallelism

This study focuses on the low-power Tegra X1 System-on-Chip (SoC) from the Jetson Nano Developer Kit, which is increasingly used in various environments and tasks. As these SoCs grow in prevalence, it becomes crucial to analyse their computational performance, energy consumption, and reliability, especially for safety-critical applications. A key factor examined in this paper is the SoC’s neutron radiation tolerance. This is explored by subjecting a parallel version of matrix multiplication, which has been offloaded to various hardware components via OpenMP, to neutron irradiation. Through this approach, this researcher establishes a correlation between the SoC’s reliability and its computational and energy performance. The analysis enables the identification of an optimal workload distribution strategy, considering factors such as execution time, energy efficiency, and system reliability. Experimental results reveal that, while the GPU executes matrix multiplication tasks more rapidly and efficiently than the CPU, using both components only marginally reduces execution time. Interestingly, GPU usage significantly increases the SoC’s critical section, leading to an escalated error rate for both Detected Unrecoverable Errors (DUE) and Silent Data Corruptions (SDC), with the CPU showing a higher average number of affected elements per SDC.

TD3-based trajectory optimization for energy consumption minimization in UAV-assisted MEC system

Unmanned Aerial Vehicle (UAV) assisted Mobile Edge Computing (MEC) systems provide substantial benefits for task offloading and communication services, especially in situations where traditional communication infrastructure is unavailable. Current research emphasizes maintaining communication quality while minimizing total energy consumption and optimizing UAV flight trajectories. However, several issues remain: First, the energy consumption objective function lacks comprehensiveness, neglecting the impact of UAV flight energy consumption; second, an effective Deep Reinforcement Learning (DRL) algorithm has not been employed to address the non-convexity of the objective function; third, there is insufficient discussion regarding the practical significance of the proposed approach. To address these issues, this paper formulates an objective function aimed at minimizing MEC energy consumption by considering task offloading decisions, communication delays, computational energy consumption, and UAV flight energy consumption. We propose a Population Diversity-based Particle Swarm Optimization-Double Delay Deep Deterministic Policy Gradient (PDPSO-TD3) algorithm to find the optimal solution, enhance UAV flight trajectories through optimized offloading decisions, ensure efficient communication, and minimize the total energy consumption of the MEC system. Furthermore, we discuss the practical applicability of PDPSO-TD3 in detail and present the proposed scheme. Experimental results demonstrate that compared to the Deep Deterministic Policy Gradient (DDPG) algorithm, for transmission delay, MEC energy consumption, UAV flight energy consumption, and User Equipments (UEs) access rate metrics. The proposed PDPSO-TD3 algorithm can improvement the performance by about 14.3%, 10.1%, 6.1%, and 3.3%, respectively.

A joint vehicular device scheduling and uncertain resource management scheme for Federated Learning in Internet of Vehicles

Federated learning (FL) offers an effective framework for the efficient process in vehicular edge computing. However, FL encompasses the process of distributing and uploading model parameters, which are inevitably transmitted in a wireless network environment. Some challenges in FL-assisted Internet of Vehicles (IoV) sceneries gradually emerging, such as data heterogeneity, concerned device resources, and unstable communication environment, which necessitate intelligent vehicle selection schemes that accelerate training efficiency. Based on these, we consider a new scenario, specifically an FL-assisted IoV system under uncertain communication conditions, and develop an interval many-objective vehicle selection and bandwidth allocation (IMoVSBA) joint optimization scheme. This scheme takes into account computation latency, energy consumption, server utilization, and data quality, while meeting multi-criteria resource optimization requirements. Among these, server utilization is a new objective designed specifically for this joint optimization problem. For the proposed problem, a novel interval many-objective evolutionary algorithm with individual comprehensive indicator to control the evolution direction (IMaOEACI) is designed. Simulation results demonstrate that this method outperforms other schemes in terms of accuracy, training cost, and server utilization, effectively improving training efficiency in wireless channel environments and reasonably utilizing bandwidth resources. It provides significant scientific value and application potential in the field of the IoVs.

Mobile Device Security: A Two-Layered Approach with Blockchain and Sensor Technology for Theft Prevention

In the backdrop of the escalating incidents of mobile device theft and associated security challenges, a resilient and innovative solution is imperative. The traditional security mechanisms, largely reliant on the International Mobile Equipment Identity (IMEI), have been fraught with vulnerabilities, leading to a surge in incidents of device theft and data breaches. Addressing this pressing issue, we present a novel, two-tiered approach integrating sensor technology and blockchain to bolster mobile device security. This work aims to create and assess a dual-layered security strategy that uses blockchain and sensor technologies in a complementary way. Based on a rigorous conceptual framework, it explores a two-tiered security model that intricately combines sensor and blockchain technologies. This collaborative integration aims to provide an effective solution to the widespread challenges of theft and security breaches in mobile devices. The methodology employs a sensor layer for real-time data collection and processing to detect potential thefts, and activating alerts. The blockchain layer, invoked upon these alerts, initiates secure, transparent, and decentralized transactions for verification and validation across network nodes. This dual mechanism ensures swift and secure anti-theft actions, supported by an enhanced encryption standard. Our result analysis reveals the proposed system’s superiority in computational time, energy consumption, and overall security levels when compared to existing protocols. The integration of real-time processing and blockchain’s immutable nature promises reduced false positives and enhanced data integrity. The findings indicate that this integrative approach not only mitigates theft but also ensures data security, marking a significant stride in mobile security technology. In conclusion, this two-layered system promises a scalable, efficient, and robust solution to mobile device theft and data breaches, with potential impacts transcending individual device security to influence broader data privacy and security paradigms, thus signifying a pivotal development in the field of mobile security.

Comparative Study of Energy Consumption Intensity: A Case Study of the Faculty of Electrical Engineering and Informatics (FEI) Versus the Faculty of Chemical Technology (FChT) in the City of Pardubice

Abstract The study’s comparative aspect between the Faculty of Electrical Engineering and Informatics (FEI) and the Faculty of Chemical Technology (FChT) buildings is pivotal for elucidating nuanced differences in energy consumption practices within academic institutions, facilitating targeted energy efficiency measures. The research methodology adopts a comparative case study approach to examine two focal research buildings. Analysis entails computing energy consumption intensity per unit area for both buildings and comparing energy consumption data to identify disparities. The need for understanding energy consumption patterns and improving energy efficiency in academic infrastructure is directly addressed through variables studied in the comparative analysis, including energy consumption comparison, annual and monthly variations, ratio, and intensity. The data indicates that the average heating intensity in the FChT building surpasses that of the FEI building by a factor of 2.3, signifying significant disparities in heating energy consumption trends between the two structures and suggesting potential areas for targeted energy efficiency interventions. Moreover, the average electrical intensity in the FChT building is over 1.7 times higher than in the FEI building. Interestingly, despite the FChT building being 2.6 times larger than the FEI, its intensity does not always correspond with the increase in building area, indicating additional factors influencing energy consumption.

Deploying large language models on diverse computing architectures: A performance evaluation framework

Deploying large language models (LLMs) across diverse computing architectures is a critical challenge in the field of artificial intelligence, particularly as these models become increasingly complex and resource-intensive. This review presents a performance evaluation framework designed to systematically assess the deployment of LLMs on various computing architectures, including CPUs, GPUs, TPUs, and specialized accelerators. The framework is structured around key performance metrics such as computational efficiency, latency, throughput, energy consumption, and scalability. It considers the trade-offs associated with different hardware configurations, optimizing the deployment to meet specific application requirements. The evaluation framework employs a multi-faceted approach, integrating both theoretical and empirical analyses to offer comprehensive insights into the performance dynamics of LLMs. This includes benchmarking LLMs under varying workloads, data batch sizes, and precision levels, enabling a nuanced understanding of how these factors influence model performance across different hardware environments. Additionally, the framework emphasizes the importance of model parallelism and distribution strategies, which are critical for efficiently scaling LLMs on high-performance computing clusters. A significant contribution of this framework is its ability to guide practitioners in selecting the optimal computing architecture for LLM deployment based on application-specific needs, such as low-latency inference for real-time applications or energy-efficient processing for large-scale deployments. The framework also provides insights into cost-performance trade-offs, offering guidance for balancing the financial implications of different deployment strategies with their performance benefits. Overall, this performance evaluation framework is a valuable tool for researchers and engineers, facilitating the efficient deployment of LLMs on diverse computing architectures. By offering a systematic approach to evaluating and optimizing LLM performance, the framework supports the ongoing development and application of these models across various domains. This paper will evaluate the deployment of large language models (LLMs) on diverse computing architectures, including x86, ARM, and RISC-V platforms. It will discuss strategies for optimizing LLM performance, such as dynamic frequency scaling, core scaling, and memory optimization. The research will contribute to understanding the best practices for deploying AI applications on different architectures, supporting technological innovation and competitiveness.

Multi-UAV Assisted Air–Ground Collaborative MEC System: DRL-Based Joint Task Offloading and Resource Allocation and 3D UAV Trajectory Optimization

In disaster-stricken areas that were severely damaged by earthquakes, typhoons, floods, mudslides, and the like, employing unmanned aerial vehicles (UAVs) as airborne base stations for mobile edge computing (MEC) constitutes an effective solution. Concerning this, we investigate a 3D air–ground collaborative MEC scenario facilitated by multi-UAV for multiple ground devices (GDs). Specifically, we first design a 3D multi-UAV-assisted air–ground cooperative MEC system, and construct system communication, computation, and UAV flight energy consumption models. Subsequently, a cooperative resource optimization (CRO) problem is proposed by jointly optimizing task offloading, UAV flight trajectories, and edge computing resource allocation to minimize the total energy consumption of the system. Further, the CRO problem is decoupled into two sub-problems. Among them, the MATD3 deep reinforcement learning algorithm is utilized to jointly optimize the offloading decisions of GDs and the flight trajectories of UAVs; subsequently, the optimal resource allocation scheme at the edge is demonstrated through the derivation of KKT conditions. Finally, the simulation results show that the algorithm has good convergence compared with other algorithms and can effectively reduce the system energy consumption.

Obfuscating Ciphertext-Policy Attribute-Based Re-Encryption for Sensor Networks with Cloud Storage

With the rapid growth of wireless sensor networks, secure data transmission, storage, and distribution in such networks has become an urgent demand. To defend against security risks such as data leakage, key compromise, and unauthorized misuse of data simultaneously, we propose a novel obfuscatable ciphertext-policy attribute-based re-encryption scheme with a specially designed obfuscator. The proposed scheme leverages program obfuscation to transform the re-encryption program codes into an unintelligible form and embed the private keys into the obfuscated implementation. Consequently, the proposed scheme protects data confidentiality and keeps the secrecy of the private key while providing fine-grained access control. Formal proofs for the security of the proposed re-encryption scheme and the obfuscator are provided. Extensive experiments have been conducted on representative platforms, including cloud servers, workstations, and embedded devices, to evaluate the computational efficiency and energy consumption of the scheme. Experimental results indicate that the scheme achieves high efficiency on various platforms and economical energy consumption on typical embedded devices with constrained resources.

Performance evaluation and analysis of public-key schemes in embedded IoT devices

The Internet of Things (IoT) is considered one of the emerging technologies that have attracted widespread attention from industry and academia as a result of their ability to use them in many applications, including military, healthcare, and industrial control. The inherent vulnerabilities in communications of IoT networks, which consist of resource-constrained devices, pose significant security challenges. To protect sensitive data exchanged within these networks, it is necessary to design and implement lightweight and efficient cryptographic schemes that consider resource-constrained IoT devices. This paper evaluates the performance of five well-known public-key schemes on three real-world IoT devices including IOTLAB-M3, Arduino-Zero, and Decawave DWM1001. The examined public-key schemes are categorized into two classes: integer factorization-based (RSA and Rabin schemes) and elliptic curve-based (ECIES, ECDH, and ECDSA schemes). By conducting comprehensive experiments, we compare these schemes in terms of computational time and energy consumption, providing valuable insights for selecting optimal public-key schemes in resource-constrained IoT environments.

An AKA protocol for 5G-assisted D2D communication in Out-of-Coverage scenario

5G-assisted Device to Device (D2D) communication can be broadly categorized into three use case scenarios: In Coverage, Relay Coverage, and Out-of Coverage. The main challenge lies in ensuring secure communication in Out-of Coverage scenarios, as in this situation, neither of the two devices is within the 5G network’s coverage area. Although several researchers have developed authentication mechanisms for securing D2D communication, most are unsuitable for Out-of Coverage scenarios. Additionally, many of these mechanisms cannot withstand free-riding attacks due to the absence of a trusted entity. In a 5G cellular network, a trust relationship can be established between a registered device and the home network through mutual authentication whenever the former is within the latter’s coverage area. Leveraging this trust, this paper proposes a lightweight 5G-assisted authentication protocol for mutual authentication between two communicating devices in Out-of Coverage scenarios. The proposed protocol meets the necessary security goals and mitigates various security attacks, including free-riding attacks. The correctness of the proposed protocol is formally established using the Scyther tool and Random Oracle Model. Furthermore, performance analysis shows that the protocol is efficient in terms of computation overhead, communication overhead and energy consumption compared to similar works in the literature. The computation overhead is found to be 596.12 ms for the Requestor device and 587.26 ms for the Requestee device. The communication overhead is 2720 bits and the total energy consumption for both devices combined is found to be 3016.93 millijoules.

Dependency-aware online task offloading based on deep reinforcement learning for IoV

The convergence of artificial intelligence and in-vehicle wireless communication technologies, promises to fulfill the pressing communication needs of the Internet of Vehicles (IoV) while promoting the development of vehicle applications. However, making real-time dependency-aware task offloading decisions is difficult due to the high mobility of vehicles and the dynamic nature of the network environment. This leads to additional application computation time and energy consumption, increasing the risk of offloading failures for computationally intensive and latency-sensitive applications. In this paper, an offloading strategy for vehicle applications that jointly considers latency and energy consumption in the base station cooperative computing model is proposed. Firstly, we establish a collaborative offloading model involving multiple vehicles, multiple base stations, and multiple edge servers. Transferring vehicular applications to the application queue of edge servers and prioritizing them based on their completion deadlines. Secondly, each vehicular application is modeled as a directed acyclic graph (DAG) task with data dependency relationships. Subsequently, we propose a task offloading method based on task dependency awareness in deep reinforcement learning (DAG-DQN). Tasks are assigned to edge servers at different base stations, and edge servers collaborate to process tasks, minimizing vehicle application completion time and reducing edge server energy consumption. Finally, simulation results show that compared with the heuristic method, our proposed DAG-DQN method reduces task completion time by 16%, reduces system energy consumption by 19%, and improves decision-making efficiency by 70%.