Energy efficiency is crucial for Connected Autonomous Vehicles (CAVs), where real-time perception via Deep Neural Networks (DNNs) demands significant computing resources. Although techniques such as Dynamic Frequency Scaling (DFS) and block-skipping reduce energy usage, they may degrade accuracy or increase inference speed. Integrating these methods with Collaborative Perception (CP), which leverages data sharing among vehicles to improve perception performances, offers a potential trade-off between accuracy and energy usage. This paper introduces EC-PUBSE (Efficient Collaborative Perception Using Block-Skipping, DFS, and Environment awareness), an energy-efficient CP framework for CAVs. EC-PUBSE dynamically allocates computing resources using block-skipping and DFS, coupled with environment-aware response time budget to sustain accuracy and reduce energy. We formulate the problem as a mixed-integer nonlinear programming (MINLP) problem, addressing the trade-off between computation energy and accuracy, under communication and response time budget constraints. We leverage model diversity and redundancy to improve system performance compared to standalone and CP. The proposed framework adapts to traffic conditions to offer an adaptable solution for sustainable autonomous driving. Experiments show that EC-PUBSE reduces energy consumption by up to 22%; it improves energy efficiency by up to$2.7\times $and$2.3\times $vs. standalone and collaborative only, respectively, maintaining perception performance in dynamic conditions.

In edge intelligence systems, deep neural network (DNN) partitioning and data offloading can provide real-time task inference for resource-constrained mobile devices. However, the inference time of DNNs is typically uncertain and cannot be precisely determined in advance, presenting significant challenges in ensuring timely task processing within deadlines. To address the uncertain inference time, we propose a robust optimization scheme to minimize the total energy consumption of mobile devices while meeting task probabilistic deadlines. The scheme only requires the mean and variance information of the inference time, without any prediction methods or distribution functions. The problem is formulated as a mixed-integer nonlinear programming (MINLP) that involves jointly optimizing the DNN model partitioning and the allocation of local CPU/GPU frequencies and uplink bandwidth. To tackle the problem, we first decompose the original problem into two subproblems: resource allocation and DNN model partitioning. Subsequently, the two subproblems with probability constraints are equivalently transformed into deterministic optimization problems using the chance-constrained programming (CCP) method. Finally, the convex optimization technique and the penalty convex-concave procedure (PCCP) technique are employed to obtain the optimal solution of the resource allocation subproblem and a stationary point of the DNN model partitioning subproblem, respectively. The proposed algorithm leverages real-world data from popular hardware platforms and is evaluated on widely used DNN models. Extensive simulations show that our proposed algorithm effectively addresses the inference time uncertainty with probabilistic deadline guarantees while minimizing the energy consumption of mobile devices.

Deep neural networks (DNNs) have been widely applied in diverse applications, but the problems of high latency and energy overhead are inevitable on resource-constrained devices. To address this challenge, most researchers focus on the dynamic voltage and frequency scaling (DVFS) technique to balance the latency and energy consumption by changing the computing frequency of processors. However, the adjustment of memory frequency is usually ignored and not fully utilized to achieve efficient DNN inference, which also plays a significant role in the inference time and energy consumption. In this paper, we first investigate the impact of joint memory frequency and computing frequency scaling on the inference time and energy consumption with a model-based and data-driven method. Then by combining with the fitting parameters of different DNN models, we give a preliminary analysis for the proposed model to see the effects of adjusting memory frequency and computing frequency simultaneously. Finally, simulation results in local inference and cooperative inference cases further validate the effectiveness of jointly scaling the memory frequency and computing frequency to reduce the energy consumption of devices.

The rapid development of deep neural networks (DNNs) is inherently accompanied by the problem of high computational costs. To tackle this challenge, dynamic voltage frequency scaling (DVFS) is emerging as a promising technology for balancing the latency and energy consumption of DNN inference by adjusting the computing frequency of processors. However, most existing models of DNN inference time are based on the CPU-DVFS technique, and directly applying the CPUDVFS model to DNN inference on GPUs will lead to significant errors in optimizing latency and energy consumption. In this paper, we propose a DVFS-aware latency model to precisely characterize DNN inference time on GPUs. We first formulate the DNN inference time based on extensive experiment results for different devices and analyze the impact of fitting parameters. Then by dividing DNNs into multiple blocks and obtaining the actual inference time, the proposed model is further verified. Finally, we compare our proposed model with the CPU-DVFS model in two specific cases. Evaluation results demonstrate that local inference optimization with our proposed model achieves a reduction of no less than 66% and 69% in inference time and energy consumption respectively. In addition, cooperative inference with our proposed model can improve the partition policy and reduce the energy consumption compared to the CPUDVFS model.

Collaborative perception (CP) is emerging as a promising solution to the inherent limitations of stand-alone intelligence. However, current wireless communication systems are unable to support feature-level and raw-level collaborative algorithms due to their enormous bandwidth demands. In this paper, we propose DiffCP, a novel CP paradigm that utilizes a diffusion model to efficiently compress the sensing information of collaborators. By incorporating both geometric and semantic conditions into the generative model, DiffCP enables feature-level collaboration with an ultra-low communication cost, advancing the practical implementation of CP systems. This paradigm can be seamlessly integrated into existing CP algorithms to enhance a wide range of downstream tasks. Through extensive experimentation, we investigate the tradeoffs between communication, computation, and performance. Numerical results demonstrate that DiffCP can significantly reduce communication costs by 14.5-fold while maintaining the same performance as the state-of-the-art algorithm.

In edge intelligence systems, task offloading and resource allocation policies critically depend on the required computing capacity of the task, which can only be accurately measured after execution, presenting significant design challenges. In this paper, we address the problem of robust task offloading and resource allocation under imperfect computing capacity information, where the exact value as well as distribution knowledge of the required computing capacity cannot be obtained in advance. Specifically, we formulate theenergy-time cost(ETC) minimization problem using min-max robust optimization. To tackle this challenging issue, we propose a decoupling method. This method first assumes the offloading policy is predetermined and derives two independent subproblems: local ETC and edge ETC. Then, we provide a closed-form optimal solution for the local ETC problem. The edge ETC problem is equivalently transformed into a geometric programming (GP) problem, and we introduce an effective iterative algorithm to obtain a stationary point, utilizing successive convex approximation (SCA). Finally, we design a coordinate descent (CD)-based algorithm to optimize the offloading policy effectively. Extensive simulations demonstrate that the proposed policy significantly outperforms other benchmark methods, achieving near-optimal performance even in the presence of high estimation errors in computing capacity.

Leveraging the computing and sensing capabilities of vehicles, vehicular federated learning (VFL) has been applied to edge training for connected vehicles. The dynamic and inter-connected nature of vehicular networks presents unique opportunities to harness direct vehicle-to-vehicle (V2V) communications, enhancing VFL training efficiency. In this paper, we formulate a stochastic optimization problem to optimize the VFL training performance, considering the energy constraints and mobility of vehicles, and propose a V2V-enhanced dynamic scheduling (VEDS) algorithm to solve it. The model aggregation requirements of VFL and the limited transmission time due to mobility result in a stepwise objective function, which presents challenges in solving the problem. We thus propose a derivative-based drift-plus-penalty method to convert the long-term stochastic optimization problem to an online mixed integer nonlinear programming (MINLP) problem, and provide a theoretical analysis to bound the performance gap between the online solution and the offline optimal solution. Further analysis of the scheduling priority reduces the original problem into a set of convex optimization problems, which are efficiently solved using the interior-point method. Experimental results demonstrate that compared with the state-of-the-art benchmarks, the proposed algorithm enhances the image classification accuracy on the CIFAR-10 dataset by 4.20% and reduces the average displacement errors on the Argoverse trajectory prediction dataset by 9.82%.

In recent years, the concept of collaborative perception (CP) in self-driving vehicles has emerged as a new paradigm for augmenting the safety and efficiency of connected autonomous vehicles (CAVs). However, CP's energy consumption remains a major concern, due to their computation- and transmission-intensive characteristics. To address this issue, this paper first presents a theoretical definition of CP coverage along with a 2-dimensional CP model, followed by a novel framework that leverages roadside units (RSU) to facilitate CP, namely the RSU-Aided Energy-Efficient Sensing, Computation, and Communication (RE2SCC). Through a mix of centralized scheduling and a decentralized data-sharing approach, RE2SCC improves perception performance and energy efficiency. The core of RE2SCC is a novel approach for reducing the overall computation load and energy-efficient CP by scheduling computation and transmission depending on CAVs topology while maintaining the perception performance. The centralized scheduling exploits CP capabilities via sensing data selection, avoiding redundant computation, and direct transmission of perception object data to CAVs, enabling extended perception while minimizing the transmission power. Simulations show the efficiency of the RE2SCC framework for energy savings along with increased perception performance by up to 51% in a given scenario.

With the rapid advancement of intelligent vehicle technology, the demand for vehicular computing power is increasing. To alleviate computing loads on the on-board computer, this paper proposes an in-vehicle distributed computing system that leverages in-vehicle devices, such as smartphones and tablet computers, for cooperative task execution. Different from previous works that focused on the impact of CPU frequency management for computation offloading, we consider the scenario where the CPU frequency of devices cannot be adjusted, and study a more practical approach to make offloading and scheduling decisions by considering the CPU utilization of devices. Therefore, we first derive an analytical relationship between CPU utilization and computation latency, and then propose a CPU Utilization-Aware Scheduling (CUAS) policy to minimize response latency consisting of computation and communication latency. Simulations conducted on Simgrid show that our proposed CUAS policy can reduce the response latency by up to 20.60% compared with the benchmarks. Additionally, we established a real-world testbed to validate our system's practicality. Experimental results indicate that our proposed policy can reduce response latency by up to 20.75% compared with the benchmarks.

The introduction of 5G has changed the wireless communication industry. Whereas previous generations of cellular technology are mainly based on communication for people, the wireless industry is discovering that 5G may be an era of communications that is mainly focused on machine-to-machine communication. The application of Ultra Reliable Low Latency Communication in factory automation is an area of great interest as it unlocks potential applications that traditional wired communications did not allow. In particular, the decrease in the inter-device distance has led to the discussion of coding schemes for these interference-filled channels. To meet the latency and accuracy requirements of URLLC, Non-orthogonal multiple access has been proposed but it comes with associated challenges. In order to combat the issue of interference, an enhanced version of Sliding window superposition coding has been proposed as a method of coding that yields performance gains in scenarios with high interference. This paper examines the abilities of this coding scheme in a broadcast network in 5G to evaluate its robustness in situations where interference is treated as noise in a factory automation setting. This work shows the improvements of enhanced sliding window superposition coding over benchmark protocols in the high-reliability requirement regions of block error rates $\approx 10^{-6}$

In this paper, we study the problem of joint Task offloading and resource Allocation for vehicular edge computing with Result Feedback Delay (TARFD). Specifically, we consider a typical roadside unit (RSU) and vehicles within its coverage area, and optimize the task offloading decisions of vehicles as well as the uplink bandwidth allocation and the computation resources allocation on the RSU. The TARFD problem is formulated as a non-convex mixed integer nonlinear programming (MINLP) to minimize the average delay consisting of task offloading delay, task computation delay, and result feedback delay. We derive a lower bound of the optimum to the TARFD problem, based on which we propose an approximate algorithm of the TARFD problem, called A-TARFD. The A-TARFD algorithm can effectively deliver solutions for small-scale scenarios. To tackle large-scale scenarios, a low-complexity algorithm for the TARFD problem, called L-TARFD, is developed by constructing an iteratively updated sequence of locally tight approximate geometric programming (GP) problems. The L-TARFD algorithm can converge to a Karush-Kuhn-Tucker (KKT) point and forces the offloading decisions arbitrarily close to binary values. By comparison with the lower bound, simulation results show that the proposed two algorithms have near-optimal performance over a wide range of parameter settings.

The rapid development of the Internet of Vehicles (IoV) enables various vehicular applications, such as image-aided navigation and traffic information management. It is important to provide efficient content delivery services for these vehicular applications. Caching popular content at roadside units (RSUs) is a promising way to improve content delivery efficiency. However, due to RSUs with limited cache space, it is very challenging to develop an effective content delivery policy that satisfies the high quality of service (QoS) requirements for vehicular applications. In this paper, we investigate the user-centric content delivery problem with service delay constraints in the IoV, where the objective is to minimize the vehicle’s cost under usage-based pricing. The problem of finding an optimal content delivery policy is modeled as a finite-horizon Markov decision process (MDP). Since the cache state of each RSU, and the wireless channel qualities between the vehicle and RSUs, are usually unknown to the vehicle a priori, the vehicle must learn the optimal delivery policy by interacting with the environment. To solve this problem and optimize the vehicle’s cost, we propose a double deep Q network (DDQN)-based algorithm, which implements dynamic content delivery decisions. Furthermore, the double deep Q network can overcome the large-scale state space and reduce Q value over-estimation. Numerical results show that our policy achieves a near-optimal performance when compared to the optimal policy that knows precisely cache state and wireless channel state. We also compare the effects of different caching strategies and vehicle mobility on the performance of the algorithm.

Caching popular contents at radio access networks is a promising approach to improve the content delivery efficiency. Most of the existing content delivery schemes focus on the perspective of content providers, paying less attention to the service demand of content requesters. In this paper, we investigate the content delivery policy of a mobile device with service delay constraint in a cache- enabled heterogeneous network (HetNet), where a macro base station (MBS) is overlaid with some small base stations (SBS) with caches. In the considered network, the mobile device needs to make content delivery decisions based on the time, cache state, and signal-to-interference-plus-noise ratio (SINR) state. The problem of solving an optimal content delivery policy is modeled as a Markov decision process (MDP), where the objective is to minimize the delivery cost of the mobile device under the constraint of content service deadline. In order to address this problem, we propose a reinforcement learning (RL) algorithm to learn the optimal policy. The simulation results demonstrate that our proposed RL-based policy achieves a significant improvement in content delivery cost compared with other benchmark solutions.