Next-generation mobile networks are expected to facilitate fast AI model downloading to end users. By caching models on edge servers, mobile networks can deliver models to end users with low latency, resulting in a paradigm of edge model caching. In this paper, we develop a novel model placement framework, called parameter-sharing model caching (TrimCaching). TrimCaching exploits the key observation that a wide range of AI models, such as convolutional neural networks or large language models, can share a significant proportion of parameter blocks containing reusable knowledge, thereby improving storage efficiency. To this end, we formulate a parameter-sharing model placement problem to maximize the cache hit ratio in multi-edge wireless networks by balancing the fundamental tradeoff between storage efficiency and service latency. We show that the formulated problem is a submodular maximization problem with submodular constraints, for which no polynomial-time approximation algorithm exists. To tackle this challenge, we study an important special case, where a small fixed number of parameter blocks are shared across models, which often holds in practice. In such a case, a polynomial-time algorithm with a $\left(1-ε\right)/2$-approximation guarantee is developed. Subsequently, we address the original problem for the general case by developing a greedy algorithm. Simulation results demonstrate that the proposed TrimCaching framework significantly improves the cache hit ratio compared with state-of-the-art content caching without exploiting shared parameters in AI models.

The increasing complexity of deep neural networks poses significant barriers to democratizing AI to resource-limited edge devices. To address this challenge, split federated learning (SFL) has emerged as a promising solution that enables device-server co-training through model splitting. However, although system optimization substantially influences the performance of SFL, the problem remains largely uncharted. In this paper, we first provide a unified convergence analysis of SFL, which quantifies the impact of model splitting (MS) and client-side model aggregation (MA) on its learning performance, laying a theoretical foundation for this field. Based on this convergence bound, we introduce AdaptSFL, an adaptive SFL framework to accelerate SFL under resource-constrained edge computing systems. Specifically, AdaptSFL adaptively controls MS and client-side MA to balance communication-computing latency and training convergence. Extensive simulations across various datasets validate that our proposed AdaptSFL framework takes considerably less time to achieve target accuracy than existing benchmarks.

Next-generation mobile networks are expected to facilitate fast AI model downloading to end users. By caching models on edge servers, mobile networks can deliver models to end users with low latency, resulting in a paradigm called edge model caching. In this paper, we develop a novel model placement scheme, called parameter-sharing model caching (TrimCaching). TrimCaching exploits the key observation that a wide range of AI models, such as convolutional neural networks or large language models, can share a significant proportion of parameter blocks containing reusable knowledge, thereby improving storage efficiency. To this end, we formulate a parameter-sharing model placement problem to maximize the cache hit ratio in multi-edge wireless networks by balancing the fundamental tradeoff between storage efficiency and service latency. We show that the formulated problem is a submodular maximization problem with submodular constraints, for which no polynomial-time approximation algorithm exists. To overcome this challenge, we study an important special case, where a small fixed number of parameter blocks are shared across models, which often holds in practice. In such a case, a polynomial-time algorithm with (1 - E) /2-approximation guarantee is developed. Subsequently, we address the original problem for the general case by developing a greedy algorithm. Simulation results demonstrate that the proposed TrimCaching framework significantly improves the cache hit ratio compared with state-of-the-art content caching without exploiting shared parameters in AI models.

Massive Internet of Things (MIoT) devices in the areas without cellular networks have difficulties transmitting data to the network. Since the vehicles have sufficient energy resources and the number of vehicles is high, vehicles passing through these areas can collect MIoT data and relay data to the cellular networks. In this paper, considering the limited energy resources of MIoT devices, we formulate an Energy Efficiency of Packet Uploading (EEPU) Maximization strategy to help MIoT devices upload more packets with less energy consumption to the vehicle. Also, considering the uncertainty of the vehicle arrival, we control the packet forwarding rate between devices to achieve the goal of MIoT device queue stability. The above optimization problem can be solved by the proposed EEPU Maximization Algorithm. Numerical results show that the energy efficiency of the proposed strategy is superior to other strategies, and our proposed strategy can allow a higher packet forwarding rate on the premise of ensuring queue stability.

Nowadays, massive Internet of Things (MIoT) devices play an essential role due to their easy deployment. In remote areas, although MIoT devices have difficulties accessing cellular networks directly, they can upload their collected information via the passing mobile carriers like vehicles. Considering the limited transmission range of MIoT devices and significant signaling overhead, it is necessary to preprocess these data before sending them to the vehicles. In this paper, we first formulate an energy-minimization problem to determine the optimal number of cluster heads and the optimal size of files while guaranteeing the transmission delay. After selecting cluster heads, a routing strategy is devised to enhance the link reliability further. The above optimization problems can be solved by the proposed Multi-Layer Clustering-Routing (MLCR) Algorithm. Numerical results prove the efficiency of the proposed MLCR Algorithm in MIoT data preprocessing and show a tradeoff problem between energy consumption and route reliability.

Recently, the vehicular ad hoc network (VANET) has attracted the attention of researchers with the development of the internet of things (IoT) and the intelligent transport system (ITS). One of the major application scenarios of the fifth generation wireless communication is massive machine type communication (mMTC). In order to aggregate the data recorded by machines, information packets need to be delivered to bureaus in the network. However, some packets are not very urgent and they don't have to be transferred by the cellular communication due to the fact that the spectrum source is scarce. With the increasing number of vehicles and the increasing computing power of on board units (OBUs), vehicle-to-vehicle (V2V) communications are better to deliver data packets. In order to transfer the packets effectively, it is important to find a reliable vehicular communication route. As the topology and the vehicles velocity change much more rapidly, the existing routing algorithms in other kinds of ad hoc networks are not suitable for the VANET. In this paper, we propose a vehicular routing algorithm based on graph theory. We consider the network situations more comprehensively and the simulation results show that the algorithm proposed is superior to the traditional routing algorithm.

Recently, the research on the vehicular ad hoc network (VANET) has been paid more attention by researchers with the quick development of the autonomous driving technology. In the VANET, vehicles can communicate with everything through the route established by routing algorithms. However, the topology of the VANET changes fast because the vehicles move fast. Also, as the number of vehicles increases, the probability of data collision and the transmission latency will also increase when communicating. Therefore, the VANET needs a stable, low‐latency, and efficient route for vehicles to communicate with each other. However, the existing routing algorithms are either unable to aggregate data or are not suitable for the large‐size VANET. In this paper, we consider the vehicle attribute information comprehensively and cluster the vehicles on the road by using the cluster algorithm we propose. We dynamically select the cluster heads at each moment according to their attribute information. We consider all kinds of nodes in the network and the vehicle nodes will communicate with each other through the cluster heads under the two‐level communicating algorithm we propose. Compared with the existing cluster routing algorithm, the algorithm we propose is much more suitable for the large‐size VANET because the cluster heads do not need a gateway to help them communicate. In the simulation part, we set some real street scenes in Simulation of Urban Mobility (SUMO) and the vehicles can move by the traffic rules like in the real world, which is more suitable for the VANET. After analysing the communication performance in Network Simulator version 2 (NS2), we can get a conclusion that the algorithm proposed is superior to the traditional routing algorithm. The route established by the algorithm we propose is much more stable and efficient. And the latency is also lower than the former.