The space-air-ground integrated network has emerged as a critical enabler to achieve high-capacity 6 G communications. However, frequent handovers between satellites and gateways, along with unbalanced gateway traffic, significantly degrade the overall transmission capacity. To address these issues, this paper proposes a balanced satellite-ground scheduling architecture based on the analytic hierarchy process, called AHP-BSA. First, three spatiotemporal parameters (interconnection time$R(t)$, transmission capacity$C(t)$, and propagation delay$D_{p}$) are defined. The rationality and consistency of AHP-BSA is proved using the spatiotemporal parameters. In the parameter calculation phase of AHP-BSA, this paper further proves the relationship between$R(t)$and outage probability, as well as between$D_{p}$and transmission efficiency. Then, a parameter optimization algorithm is designed in AHP-BSA. These spatiotemporal parameters are normalized, weighted, and incorporated into the parameter optimization algorithm. Through stability-aware adjustment, it adjusts scheduling decisions in response to link dynamics and real-time fluctuations. Simulation results confirm that AHP-BSA outperforms existing methods in transmission capacity, time complexity, and long-term traffic balance.

Facial Expression Recognition (FER) has recently plays a crucial role in advancing human-computer interaction systems, aiming to understand users' inner states and underlying intentions. However, FER in real-world scenarios remains challenging due to significant label noise, caused by ambiguous facial expressions in low-quality images and annotation bias. To tackle this issue, this paper proposes a novel framework, Pseudo-Entropy Guided Complementary Learning (PEGCL), designed to robustly handle noisy labels by leveraging complementary information, which trains networks using all complementary labels defined as “facial expression images that do not belong to complementary emotion labels.” This approach effectively utilizes non-target emotion labels to mitigate the impact of label noise, rather than relying solely on annotated emotion labels. Specifically, the proposed PEGCL framework consists of three components: logit normalization to stabilize predicted probabilities and prevent gradient explosions, transformed complementary learning to redistribute the optimization focus across complementary categories by leveraging pseudo-entropy guided, and random complementary label dropping to dynamically exclude subsets of complementary labels, enhancing generalization and preventing overfitting. These components collectively ensure robust and efficient optimization under noisy label conditions. Importantly, the proposed PEGCL does not require explicit noise estimation or complex label correction mechanisms, making it a simple and effective solution for real-world FER tasks. Extensive experiments on benchmark FER datasets demonstrate that PEGCL consistently outperforms existing methods, achieving the state-of-the-art robustness against label noise while maintaining high classification accuracy.

In agricultural production, fixed-wing UAVs are widely used for image acquisition of pest detection because of their long endurance. However, due to its characteristic of being constrained by the turning radius, it causes the long turning path and much energy consumption. Thus, this article designs an algorithm based on turning span selection (TSS) for fixed-wing UAV, which plans a coverage path with the shortest turning path as possible. First, the target region model of the farmland is established. Then, by analyzing the relationship between turning radius and turning span of fixed-wing UAV, three different turning strategies are proposed. Finally, according to the pointer network model and the actor-critic algorithm in reinforcement learning, the TSS algorithm is designed. Simulation results show that the proposed TSS algorithm can effectively plan a flight path with a shorter turning path, and has obvious performance improvement compared with the existing algorithms

A large number of plug-in hybrid electric vehicles (PHEVs) have high mobility but a small battery capacity; thus, these vehicles urgently need to make charging and discharging decisions in real time. This paper proposes a new architecture based on fog computing for an Internet of Vehicles energy trading system, which we call a vehicle-mounted energy fog. This architecture includes a fog computing energy center (FCEC), which manages local energy trading and reduces the peak load energy trading for an external public energy company. We model the optimization problems for energy trading under two different types of FCECs: 1) a nonprofit-driven FCEC whose goal is solely to benefit the PHEV charging and discharging operations and 2) a profit-driven FCEC whose goal is to maximize its own profits while still guaranteeing that each PHEV achieves a non-negative utility. We also propose efficient algorithms for these two types of FCECs to seek optimal pricing and make supply demand decisions. Simulation results show that our proposed algorithms are superior to existing algorithms in terms of the convergence rate, the final objective value and the evenness of the Pareto solution set. Specifically, the evenness of the Pareto solution set is improved by 23% compared to the results of the existing algorithm.

With the development of vehicular ad hoc networks, the shortage of resources is becoming increasingly serious. To alleviate this problem, many researchers have suggested the use of parked vehicles, but these proposals do not consider that parked vehicles cannot continue charging when they are turned off; this problem is termed the limited energy problem. In this paper, we consider the use of parked vehicles as relay nodes and introduce an optimal method of enabling parked vehicles to provide services in the most energy-efficient manner. The core method is divided into two steps: 1) the clustering of moving vehicles based on their communication coverage and the selection of parked vehicles between clusters as relay nodes to ensure communications among driving vehicles and 2) the dynamic use of external environmental factors to achieve energy conservation. Simulation results show that the proposed energy-saving method achieves obvious improvements compared with other parked-vehicle-based methods.

Service Function Chain (SFC) is not only helpful for saving the capital expenditure (CAPEX) and operational expenditure (OPEX) of network provider, but also can reduce energy consumption in the substrate network. However, to best of our knowledge, few researches focus on the problem of energy consumption for provisioning SFC requests in multi-domain networks. In this paper, we firstly formulate the problem of energy-efficient online SFC request provisioning across multiple domains by using integer linear programming (ILP). Then we propose a heuristic algorithm called EE-SFCO-MD for efficiently solving this problem. We conduct simulation experiments for evaluating the performance of our algorithm. The simulation results show that EE-SFCO-MD performs better than existing approaches.

Generally, a service request must specify its required virtual network functions (VNFs) and their specific order, which is known as the service function chain (SFC). When mapping SFCs, network providers face many challenges due to the requirements of maintaining the correct order and satisfying other constraints of VNFs. Furthermore, SFC orchestration becomes a more difficult problem when considered in multi-domain networks, because the confidentiality of the topology information of each domain must be considered. In this paper, we study the problem of SFC orchestration across multiple domains. We first use the full mesh aggregation approach to construct an abstracted network to guide the orchestration process, and then propose two efficient methods for SFC partitioning. Based on the SFC partitioning results, we also propose two heuristic algorithms for deploying the sub-chains in multiple domains. Moreover, when the partitioning results cannot be mapped completely, a feedback mechanism is used to repartition the SFC and improve the success ratio of orchestrating the SFC. Finally, to save bandwidth resources, we further improve our heuristic algorithms by migrating the deployment position of VNFs. The simulation results demonstrate that our proposed algorithm achieves better performance compared to existing solutions.

With the development of cloud computing, virtual machine migration is emerging as a promising technique to save energy, enhance resource utilizations, and guarantee Quality of Service (QoS) in cloud datacenters. Most of existing studies on the virtual machine migration, however are based on a single virtual machine migration. Although there are some researches on multiple virtual machines migration, the author usually does not consider the correlation among these virtual machines. In practice, in order to save energy and maintain system performance, cloud providers usually need to migrate multiple correlated virtual machines or migrate the entire virtual datacenter (VDC) request. In this paper, we focus on the efficient online live migration of multiple correlated VMs in VDC requests, for optimizing the migration performance. To solve this problem, we propose an efficient VDC migration algorithm (VDC-M). We use the US-wide US National Science Foundation (NSF) network as substrate network to conduct extensive simulation experiments. Simulation results show that the performance of the proposed algorithm is promising in terms of the total VDC remapping cost, the blocking ratio, the average migration time and the average downtime.

Location-Based Service (LBS) is one of the fundamental and central functionalities of mobile social networks. Since users usually have to report their locations to the LBS providers while using services, the protection of user's location privacy poses a critical challenge. Although many existing approaches can preserve user's location privacy effectively, most of them must include and use the user's real location. In this paper, we first propose an efficient k-anonymity based Dummy Location and divided Circular Area (k-DLCA) approach to protect the user's location privacy. Different from existing studies, the k-DLCA algorithm adopts a greedy strategy to select dummy locations and considers the semantic location information of the location. Moreover, the user's real location may not be contained in the chosen dummy locations. We then show that k-DLCA algorithm can resist the attacks from adversaries, and has a low probability of exposing the user's real location. We conduct extensive simulations to evaluate the efficiency of the proposed scheme. The simulation results demonstrate that our proposed scheme is promising.

Preserving user location and trajectory privacy while using location-based service (LBS) is an important issue. To address this problem, we first construct three kinds of attack models that can expose a user's trajectory or path while the user is sending continuous queries to a LBS server. Then we propose the k- anonymity trajectory (KAT) algorithm, which is suitable for both single query and continuous queries. Different from existing works, the KAT algorithm selects k-1 dummy locations using the sliding widow based k- anonymity mechanism when the user is making single queries and selects k-1 dummy trajectories using the trajectory selection mechanism for continuous queries. We evaluate and validate the effectiveness of our proposed algorithm by conducting simulations for the single and continuous query scenarios.

Virtual Network Mapping (VNM) has been a key issue for network virtualization need to be addressed. A traditional substrate network is usually managed by multicast InPs, and many applications in the substrate network can be characterized by hybrid virtual networks with both unicast and multicast requests. However, to the best of our knowledge, few researches focus on the problem of hybrid virtual network mapping across multiple domains (HVNMMD). In this paper, we investigate the HVNMMD problem and propose two algorithms to solve this problem efficiently: i) the decomposition-based algorithm, HVNMMD_D; and ii) the spectral clustering based algorithm, HVNMMD_SC. We evaluate the performance of our algorithms through conducting simulation experiments. The simulation results show that the proposed algorithms perform better than existing approaches.

One of the challenge issues in network virtualization is the efficient mapping of a virtual network (VN) onto a shared substrate network. The VN mapping problem has been addressed by various researchers. However, these solutions and associated algorithms are only efficient for building unicast service oriented virtual networks, and are not applicable to multicast service oriented VNs. Furthermore, how to guarantee survivability while provisioning a virtual multicast service oriented network is an important issue that has not been addressed. In this work, we investigate the survivable multicast service oriented virtual network provisioning(SMVNP) problem and propose an efficient algorithm with resource sharing for solving this problem. We validate and evaluate our framework and algorithms by conducting simulations on realistic substrate network and by comparing with existing approach. Our simulation results show that our approach outperforms existing solution.

Cloud based datacenters provide on demand services and resources, both transparently and cost effectively. These services and applications are typically hosted and run on the servers located in interconnected datacenters. The task or application request from users can be abstracted as a virtual network (VN) request. How to efficiently accommodate VN requests by mapping them onto the substrate network is a challenging problem. Current research only considers static VN requests, where the VN request and the demand for resources is fixed and does not change over time. However, most of the application requests submitted to datacenters present dynamic changing characteristics. In this paper, we address the issue of how to optimally reconfigure and map an existing VN while this VN request changes. As the VN provisioning problem is NP-hard, we propose heuristic algorithms for solving it efficiently. We evaluate the performance and effectiveness our algorithms by conducting simulations on realistic network. The simulation results show that our approach solves this problem efficiently and performs well than existing approaches.

In the cloud computing paradigm users access applications/services hosted and run on virtual machines in interconnected datacenters. Applications from the same user may need to interact and change data or information, thus, we may abstract the applications/services request from same user as a virtual network (VN). To improve performance and resource efficiency, it is critical that the VN request be optimally provisioned given the current resource state of the datacenters. However, all of the existing research about optimal VN provisioning conducted for deterministic services. In this work, we formulate the problem of optimal provisioning for elastic service oriented VN request as a mixed integer programming (MIP) model with the objective of maximizing the total revenue of infrastructure provider (InP), and propose a genetic algorithm based heuristic (GAH) algorithm for addressing the optimal provisioning for virtual network request with un-splittable flow (OPVNUF) and optimal provisioning for virtual network request with splittable flow (OPVNSF) problems of elastic services. We demonstrate the effectiveness of our approach for improving the total revenue, by conducting extensive simulations on real substrate network.

In P2P file sharing systems over mobile cellular networks, the bottleneck of file transfer speed is usually the downlink bandwidth of the receiver rather than the uplink bandwidth of the senders. In this paper we consider the impact of downlink bandwidth limitation on file transfer speed and propose two novel peer selection algorithms named DBaT-B and DBaT-N, which are designed for two different cases of the requesting peer's demand respectively. Our algorithms take the requesting peer's downlink bandwidth as the target of the sum of the selected peers' uplink bandwidth. To ensure load balance on cells, they will first choose a cell with the lowest traffic load before choosing each peer. We also provide a Fuzzy Cognitive Map that can be used to estimate peers' service ability in P2P systems over mobile cellular networks. Simulation results show that in respective cases DBaT-B and DBaT-N can both achieve much better load balance on cells than some traditional algorithms while ensuring favorable file transfer speed.

We propose a novel segmentation algorithm TSESS and a novel chunk scheduling algorithm TSESCS for SVC based P2P streaming systems. TSESS splits each layer into equal sized chunks and tags each chunk with a time-stamp. TSESCS is based on TSESS and adopts two scheduling windows with adaptive strategies. Compared with traditional segmentation algorithm and chunk scheduling algorithms, our TSESS and TSESCS algorithms can achieve better video quality on clients in heterogeneous network environments.

Due to the characteristics of the DCF model in 802.11 and P2P users' requirements of both downloading and uploading, P2P traffic frequently causes congestion in WLAN. However, traditional P2P cache cannot alleviates the congestion caused by P2P traffic in WLAN effectively. In this paper, we propose a novel solution called bidirectional cache for this problem. A bidirectional cache which contains a reverse cache as well as a forward cache can be deployed at the AC of a WLAN. Being a novel design, the reverse cache can cache release the uplink bandwidth resource at each AP can be and the congestion can be alleviated effectively. Meanwhile, the forward cache can still act as the traditional P2P cache to reduce the cross domain traffic. Simulation results show that, compared with traditional P2P cache, our bidirectional cache can shorten the file transfer time of P2P applications by at least 42% and improve the throughput of the other Internet applications by at least 28%.

In this paper, we focus on the problem with the objective to maximize the system performance, while guaranteeing multiple QoS (quality of service) constraints for wireless data networks accommodating multiclass services with different quality requirements. First, we formulate and solve the opportunistic scheduling problem with multiple general long-term QoS constraints. Then, we generalize this problem to include short-term QoS constraints for real-time multimedia users and long-term QoS constraints for non-real-time data users simultaneously in multiclass services networks. Simulation results illustrate that the proposed scheduling schemes guarantee the different QoS constraints, and achieve high system performance.

In this paper, we address the problem of resource allocation with efficiency and quality of service (QoS) support in uplink for a wireless CDMA network supporting real-time (RT) and nonreal-time (NRT) communication services. For RT and NRT users, there are different QoS requirements. We introduce and describe a new scheme, namely the traffic aided uplink opportunistic scheduling (TAUOS). While guaranteeing the different QoS requirements, TAUOS exploits the channel condition to improve the system throughput. In TAUOS, the cross-layer information, file size information, is used to improve the fairness of NRT users. Extensive simulation results show that our scheme can achieve high system throughput in uplink wireless CDMA system, while guaranteeing the QoS requirements.