Understanding mobile traffic patterns of large scale cellular towers in urban environment is extremely valuable for Internet service providers, users, and government managers modern metropolis. This paper aims at extracting modeling the deployed a metropolitan city. To achieve this goal, we need to address several challenges, including lack appropriate tools processing measurement data, unknown patterns, as well handling complicated factors ecology human behaviors that affect patterns. Our...

As smartphones have become indispensable personal devices, the number of smartphone users has increased dramatically over last decade. These which are supported by a variety apps, allow people to access Internet services in convenient and ubiquitous manner. App developers service providers can collect fine-grained app usage traces, revealing connections between users, smartphones. We present comprehensive review most recent research on analysis this survey. Our survey summarizes advanced...

Mobile user traffic facilitates diverse applications, including network planning and optimization, whereas large-scale mobile is hardly available due to privacy concerns. One alternative solution generate data for downstream applications. However, existing generation models cannot simulate the multi-scale temporal dynamics in on individual aggregate levels. In this work, we propose a hierarchical generative adversarial (MSH-GAN) containing multiple generators multi-class discriminator....

Understanding mobile traffic patterns of large scale cellular towers in urban environment is extremely valuable for Internet service providers, users, and government managers modern metropolis. This paper aims at extracting modeling the deployed a metropolitan city. To achieve this goal, we need to address several challenges, including lack appropriate tools processing measurement data, unknown patterns, as well handling complicated factors ecology human behaviors that affect patterns. Our...

Online services are playing critical roles in almost all aspects of users' life. Users usually have multiple online identities (IDs) different services. In order to fuse the separated user data for better business intelligence, it is service providers link IDs belonging same user. On other hand, popularity mobile networks and GPS-equipped smart devices provided a generic way IDs, i.e., utilizing mobility traces IDs. However, linking based on their has been challenging problem due highly...

Live migration is a key technique for virtual machine (VM) management in data center networks, which enables flexibility resource optimization, fault tolerance, and load balancing. Despite its usefulness, the live still introduces performance degradations during process. Thus, there has been continuous efforts reducing time order to minimize impact. From network's perspective, determined by amount of be migrated available bandwidth used such transfer. In this paper, we examine problem how...

Satellite imagery depicts the earth's surface remotely and provides comprehensive information for many applications, such as land use monitoring urban planning. Existing studies on unsupervised representation learning satellite images only take into account images' geographic information, ignoring human activity factors. To bridge this gap, we propose using Point-of-Interest (POI) data to capture factors design a contrastive learning-based framework consolidate of with POI information. Also,...

As power consumption of the Internet has been growing quickly in recent years, saving energy become an important problem networking research, for which most promising solution is to find minimum-power network subsets and shut down other unnecessary devices links satisfy changing traffic loads. However, traditional networks, it difficult implement a coordinated strategy among due their distributed control. On hand, new paradigm-software defined (SDN) provides us efficient way having...

As the volume of mobile traffic has been growing quickly in recent years, reducing congestion networks become an important problem networking research. Researchers found out that inhomogeneity spatio-temporal distribution data leads to extremely insufficient utilization network resources. Thus, it is fundamentally understand this help us make better resource planning or introduce new management tools such as time-dependent pricing reduce congestion. However, due requirement a large dataset,...

With the rapid development of mobile communication technology, trajectories humans are massively collected by Internet service providers (ISPs) and application (ASPs). On other hand, rising paradigm knowledge graph (KG) provides us a promising solution to extract structured "knowledge" from massive trajectory data. In this paper, we focus on modeling users' spatio-temporal mobility patterns based techniques, predicting future movement extracted multiple sources in cohesive manner....

Human daily activities, such as working, eating out, and traveling, play an essential role in contact tracing modeling the diffusion patterns of COVID-19 pandemic. However, individual-level activity data collected from real scenarios are highly limited due to privacy issues commercial concerns. In this paper, we present a novel framework based on generative adversarial imitation learning, generate artificial trajectories that retain both fidelity utility real-world data. To tackle inherent...

Daily activity data that records individuals' various types of activities in daily life are widely used many applications such as scheduling, recommendation, and policymaking. Though with high value, its accessibility is limited due to collection costs potential privacy issues. Therefore, simulating human produce massive high-quality great importance benefit practical applications. However, existing solutions, including rule-based methods simplified assumptions behavior data-driven directly...

Vessel trajectory prediction is the key to maritime applications such as traffic surveillance, collision avoidance, anomaly detection, and so on. Making predictions more precisely requires a better understanding of moving trend for particular vessel since movement affected by multiple factors like marine environment, type, behavior. In this paper, we propose model named VesNet, based on attentional seq2seq framework, predict future sequence observing current trajectory. Firstly, extract...

While large language models (LLMs) present significant potential for supporting numerous real-world applications and delivering positive social impacts, they still face challenges in terms of the inherent risk privacy leakage, hallucinated outputs, value misalignment, can be maliciously used generating toxic content unethical purposes after been jailbroken. Therefore, this survey, we a comprehensive review recent advancements aimed at mitigating these issues, organized across four phases LLM...

Trajectory data play a crucial role in many applications, ranging from network optimization to urban planning. Existing studies on trajectory are task-specific, and their applicability is limited the specific tasks which they have been trained, such as generation, recovery, or prediction. However, potential of unified model has not yet fully explored modeling. Although various differ inputs, outputs, objectives, conditions, share common mobility patterns. Based these patterns, we can...

Modeling information diffusion on social networks can be used to guide the prediction and control of propagation improve structure functionality networks. Existing methods predict paths its volume by modeling network user behavior. However, none existing take activity level, which is proved critical in process, into account, thus weaken accuracy. To solve this problem, paper proposes a Multi-Scale Activity Network (MSA-Net) capture topological historical affect features for different scales...

Cross-domain recommendation is a typical solution for data sparsity and cold start issue in the field of location recommendation. Specifically, an auxiliary domain leveraged to improve target domain. There scenario that two interaction domains (location based check-in service, example) combine perform cross-domain task. Existing approaches are on assumption from can be directly shared across domains. However, such not reasonable, since real world those may operated by different companies....

Network traffic data facilitates understanding the Internet of Things (IoT) behaviors and improving IoT service quality in real world. However, large-scale is rarely accessible, privacy issues also impede realistic sharing even with anonymous personal identifiable information. Researchers propose to generate synthetic but fail cover multiple services provided by widespread real-world devices. In this work, we take first step via a knowledge-enhanced generative adversarial network (GAN)...

With the rapid development of cellular network, network planning is increasingly important. Generating large-scale urban traffic contributes to via simulating behaviors planned network. Existing methods fail in long-term temporal while cannot model influences environment on networks. We propose a knowledge-enhanced GAN with multi-periodic patterns generate based environment. First, we design simulate and aperiodic dynamics learning daily patterns, weekly residual between periodic step by...