Connected Autonomous Vehicles (CAVs) are equipped with an array of sensors generating substantial data streams whose real-time analysis often exceed the onboard computational capabilities. While offloading these computational tasks to edge servers is an established solution, such an approach is increasingly challenging as multiple sensor streams need to be fused and analyzed, and thus transferred over capacity-limited and volatile wireless channels. To address this challenge, in this paper we propose a "split computing" framework. Compared to existing solutions that focus on individual sensor streams (and most commonly cameras), our framework is designed for sensor fusion neural models, and specifically camera and LiDAR fusion for semantic segmentation. Our proposed framework optimizes data transfer by eliminating pooling indices in favor of sending LiDAR point cloud indices only to the final network block. We remark how sensor fusion is instrumental to guarantee robust operations in a broad range of conditions. By compressing the data to be transported over the channel, our approach reduces offloading latency, better utilizes CAV computational resources compared to full offloading schemes and decreases channel load in congested urban network scenarios. Our experimental results demonstrate up to 62.74% improvement in task execution latency for sensor fusion models, with at most 6.61% performance trade-off due to compression.

In this paper, we focus on the scenario with offloading of computationally intensive tasks with delay constraints from users equipment (UEs) to multi-access edge computing (MEC) servers. To avoid user's dissatisfaction with offered quality of service, the computing resources should be able to handle even peak hours. As a result, a dense deployment of MEC servers should be considered in order to bring sufficient computing resources close to the UEs, thus enabling a low delay services. However, at the same time, the dense deployment of powerful MEC servers results, among others, in a high energy consumption. In this paper, we address the high energy consumption problem via a smart sleeping of the MEC servers while preserving quality of service for the UEs. To this end, we determine a set of MEC servers that should stay active and provide computation resources for the offloaded tasks while still meeting UEs requirements on delay. We formulate the problem of selecting the MEC server that can be set into sleep mode to save energy as a minimum set cover problem. Then, we propose a solution to minimize the energy consumption based on branch-and-bound algorithm to activate the MEC servers for computation ensuring the UEs requirement on delay. The effectiveness of the proposed solution is demonstrated through simulations showing that the proposal allows to save up to 34.9% of energy compared to state-of-the-art works while even slightly improving the ratio of offloaded tasks processed within required delay.

In response to the escalating demand for high-performance applications in Connected Autonomous Vehicles (CAVs), there has been an increased effort to accelerate Computer Vision (CV) tasks. Innovations in the field suggest that partitioning Deep Learning (D L) models, which are pivotal for CV tasks, can effectively utilize the computational strengths of both CAVs and proximal edge servers. This is achieved by splitting the DL models into two parts, where execution begins on the CAV, then the intermediate results are offloaded to complete processing on the edge server. Moreover, the precision and response time of these applications vary significantly, necessitating the adoption of early exit strategies. These strategies involve modifying DL models with auxiliary branches that allow for early termination at various stages, enabling the model to exit processing early when application requirements are met. To navigate the trade-offs between accuracy and latency, it is imperative to choose the model's optimal split and exit point that align with the current environmental conditions, such as server data throughput and computational load. To achieve this, we employ a state-of-the-art method that utilizes the Deep Deterministic Policy Gradient (DDPG) algorithm, augmented by Long Short-Term Memory networks. Additionally, to refine the training of the DDPG's actor network, Imitation Learning (IL) is integrated. IL employs a heuristic exhaustive search to retrospectively identify the most effective model partition and exit point.

The metaverse is a virtual space that combines physical and digital elements, creating immersive and connected digital worlds. For autonomous mobility, it enables new possibilities with edge computing and digital twins (DTs) that offer virtual prototyping, prediction, and more. DTs can be created with 3D scene reconstruction methods that capture the real world's geometry, appearance, and dynamics. However, sending data for real-time DT updates in the metaverse, such as camera images and videos from connected autonomous vehicles (CAVs) to edge servers, can increase network congestion, costs, and latency, affecting metaverse services. Herein, a new method is proposed based on distributed radiance fields (RFs), multi-access edge computing (MEC) network for video compression and metaverse DT updates. RF-based encoder and decoder are used to create and restore representations of camera images. The method is evaluated on a dataset of camera images from the CARLA simulator. Data savings of up to 80% were achieved for H.264 I-frame - P-frame pairs by using RFs instead of I-frames, while maintaining high peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM) qualitative metrics for the reconstructed images. Possible uses and challenges for the metaverse and autonomous mobility are also discussed.

With the advent of connected and autonomous vehicles (CAVs), we observe a growing need for new resource allocation solutions in mobile networks. Currently, most of the resource allocation solutions for CAVs communication do not consider the driving routes of the cars. In this paper, we introduce joint vehicular route selection and radio and computing resource allocation for CAVs. The proposed approach is based on the graph search-based lexicographic A* algorithm that minimizes the ratio of failed tasks along the entire vehicular route considering the availability of both radio and computing resources. To manage the allocation of resources among multiple CAVs for each vehicular route, we develop a blockchain-based framework allowing resource reservation by means of nonfungible tokens (NFTs). Each NFT represents an exclusive right to the required amount of radio and computing resources for the given road segment and defined time interval. The effectiveness of the proposed approach is demonstrated by simulations showing that the proposed vehicular route selection algorithm reduces the ratio of tasks not completed before the deadline by up to 69% compared to the existing state-of-the-art algorithms.

The paper presents a novel vision on the application of blockchain technology to empower the dynamic service provisioning in future 6G mobile networks. We propose a platform for decentralized service level agreement (SLA) negotiation between users and mobile network operators (MNOs) based on smart contracts and cryptocurrencies. In addition, the new quality of experience (QoE) model is proposed for end-users to customize their trade-off between SLA and service price. Finally, we develop the method of dynamic service selection among multiple MNOs that provides border-less connectivity for end-users with the guaranteed QoE regardless of the serving MNO.

Cognitive radio (CR) technology offers the possibility of an increase in spectrum utilization efficiency to resolve the prevalent spectrum scarcity problem. The economic survival of secondary spectrum markets (SSMs) is heavily dependent on the sharing of both the licensed spectrum and spectrum infrastructure by primary licensed operators (PLOs). In this research, an automated pricing model using a blockchain token called the spectrum dollar has been implemented for secondary radio spectrum trade. The use of spectrum dollars enables noncash-based secondary spectrum trade among PLOs based on a floor-and-trade rule. The pricing of spectrum dollars and the associated revenue shares are based on the underlying secondary spectrum trading behaviours of PLOs. PLOs that do not contribute enough secondary spectra to the SSM (to satisfy demand) suffer a loss proportional to the difference between their earned revenues and the specified floor value in the SSM. The secondary spectrum trade is assumed to be centrally managed by a spectrum broker, which announces the floor value for each bidding period while ensuring nonnegative revenue for the market itself. The use of the spectrum dollar along with the floor-and-trade methodology eliminates the possibilities for economic malpractice by PLOs that could increase spectrum reuse costs. In addition, the floor value provides automatic regulatory control to ensure the economic viability and prevent the technological hijacking of future SSMs.