Edge AI inference is an important workload in 5G networks. Multiple classes of edge AI workloads often share edge infrastructure, yet each may require distinct latency and rate guarantees. 5G network slicing supports differentiated requirements on the network side, but shared GPU inference also needs compute-side guarantees after requests reach the Multi-access Edge Computing (MEC) host. We extend 3GPP network slicing with compute-side enforcement so that slice guarantees remain effective after traffic reaches the MEC host. To realize this extension, we design a GPU scheduler that combines Hierarchical Token Bucket (HTB)-based traffic conditioning with Earliest Deadline First (EDF) scheduling. Our scheduler enforces per-class assured goodput, defined as the committed rate of latency-compliant completions for each class. The GPU scheduler identifies request classes via tags, which are assigned during GTPU encapsulation at the 5G user plane. This integration preserves overall latency guarantees across both the network and compute domains of a 5G slice for AI.

In heterogeneousWireless Rechargeable Sensor Networks (WRSNs) comprising both fixed and mobile nodes, existing energy replenishment strategies face severe challenges in cost, efficiency, and scalability. Fixed sensor clusters require efficient area-based charging to maintain network density, while dynamic sensors with unpredictable trajectories demand highly responsive on-demand service. Conventional single-modal or simplistic hybrid schemes cannot simultaneously satisfy these distinct and often conflicting operational requirements, leading to resource mismatch and critically limiting the overall network lifetime and quality of service. This paper introduces an Adaptive Co-operative Dual-Mode Charging Scheduling (AC-DCS) scheme, which integrates two novel algorithms. For fixed sensor clusters, the Portable-Charger Deployment Planning (PCDP) algorithm optimizes “one-to-many” charging using portable directional modules; for dynamic sensors, the Deep Adaptive Rescue Selector (DARS), a lightweight reinforcement learning model, provides ”one-to-one” on-demand service. Extensive simulations validate that in a 500-node network, AC-DCS improves the sensor survival rate by over 23%, improves mobile energy efficiency by approximately 65%, and reduces charging latency by at least 33% compared to state-of-the-art benchmarks, offering a robust solution for long-term large-scale IoT deployments.

Wireless Rechargeable Sensor Networks (WRSNs) extend the operational lifetime of battery-powered sensor deployments through Wireless Power Transfer (WPT). Traditional approaches often address node clustering, charging strategies, and charger routing independently, frequently failing to adapt to real-time demand variations and network survivability. This paper proposes a two-phase Sector-based Charging with Resilient Scheduling (SCRS) scheme designed for WRSNs utilizing fixed directional chargers. In Phase I, the SCRS scheme provides initial energy to mitigate the risk of early sensor outages. Phase II employs a quantile-based strategy to dynamically allocate charging time, thereby tightly constraining energy overflow. By optimizing charging schedules based on real-time demands, SCRS enhances energy utilization and improves network survivability. Simulation results demonstrate that SCRS reduces charging time by up to 62.96%, decreases energy overflow by up to 88.89%, and achieves near-zero sensor outage durations on average across broad configurations. Furthermore, the proposed scheme completely eliminates outages under specific high-agility configurations (e.g., utilizing the Q1 setting). Crucially, the validation context of this study incorporates rich stochastic radio models, including Nakagami-mfading and log-normal shadowing, to closely emulate realistic multipath and obstructed environments. Consequently, while the analytical bounds and comparative advantages of the SCRS scheme are robust within these environments, transitioning these findings to highly mobile real-world IoT deployments will necessitate further empirical validation in physical testbeds.

This work addresses congestion attacks on prioritization and preemption signal applications (PPSA), an important variant of cooperative intelligent transport systems (C-ITS). In PPSA, priority vehicles, such as buses and ambulances, can request traffic signal adjustments in their favor to reduce travel time. These systems rely on vehicle-to-everything (V2X) communications, which makes them vulnerable to congestion attacks. When under attack, these systems experience disruptions in information exchange between vehicles and infrastructure, leading to traffic congestion and safety hazards. Nevertheless, existing solutions do not address the attacks at a system level and within the PPSA context. Our proposal presents a software framework designed to enhance both sustainability and safety during congestion attacks. Specifically, it introduces a whitelist-based traffic-filtering mechanism that preserves system sustainability by ensuring only legitimate priority requests are processed. Additionally, a monitoring mechanism evaluates the PPSA service rate at runtime to identify and address potential safety risks arising from insufficient service rates. Moreover, we analyze the limitations of a machine learning-based defense method and highlight risks of potential service unavailability due to false positives. We believe that this research contributes to the development of a secure and safe traffic management system for smart cities.

This paper presents a “Microservices-based Agile Campus Surveillance Platform (MACSP)”, which aims to augment the existing Intelligent Surveillance system, CK-Fence, in order to solve the problems of high hardware resource usage and long alert latency. The MACSP has made significant progress and succeeded in reducing the memory usage and the system latency by 61.33% and 81.33%, respectively. The microservices architecture allows for step-by-step implementation and integration of various requirements into MACSP, making it scalable, extensible and reliable. It can be applied not only to campuses but can also be extended to other hazardous environments, such as urban and public area scenarios, making it a comprehensive solution that meets the ever-changing requirements in the field of Intelligent Surveillance.

With the explosive growth of mobile data traffic, roadside-unit (RSU) caching is considered an effective way to offload download traffic in vehicular ad hoc networks (VANETs). Many existing works investigate the content placement of RSU caching. However, few of them consider the download deadline constraint when caching the content in the RSUs. In this paper, the main objective is to maximize the hit rate of downloading the requested content from the RSUs before the deadline expires. We propose a Markov-based mobility model and a Markov clustering-based content placement algorithm to group the RSUs into clusters and allocate the content to the cache of the RSUs in the cluster. We also investigate the impact on the cache hit rate under different simulation parameters, such as the total number of RSUs, the cache size, and the number of RSUs visited by vehicles during the download period. According to the simulations conducted, when the region of interest (RoI) is small, the MVP method increases the cache hit rate by at least 21.40% compared to the existing methods. When the RoI is large, our approach outperforms other existing methods by at least 26.16% and at most 337.77%, which significantly increases the efficiency of the download session in VANET.

As the number of autonomous cars grows explosively, it is important to broadcast messages with less time and fewer transmissions, especially emergency messages. Therefore, we present an efficient way to broadcast messages by recording the information of informed and uninformed nodes and adopt the mechanism into the push-pull algorithm for vehicular ad hoc networks.

In software-defined networking (SDN), emerging new-flow attacks aim at exhausting the resources of switches and controllers through massive packet-in messages. To detect new-flow attacks, SDNGuardian was proposed as a protocol-independent defense method, which uses entropy to detect anomalies and mitigate attacks with rate limits. In this paper, we introduce a crafty new-flow attack named timeout-aware attack that SDNGuardian cannot detect. We, therefore, propose a novel defense method: SDNSnapshot. Through simulations, we show that SDNSnapshot can successfully detect the timeout-aware attack. The number of dropped benign packet-in messages in SDNSnapshot is around one third of that in SDNGuardian. Besides, a snapshot only consumes 0.9Mb static random access memory (SRAM) for each anomalous sensitive field. The results indicate that SDNSnapshot is a feasible solution to mitigate new-flow attacks in practice.

In modern world, Domain Name System (DNS) is one of the major components that build the Internet as we know today. A DNS server with geolocation capability can answer a client depending on the geolocation of the client. Unfortunately, popular DNS server-side software solutions do not support the geolocation feature. In this paper, we design a DNS server, which has geolocation feature, and implemented it. Experimental results show that we can return different values based on the geolocation information of the client, so different clients may connect to the server that are nearest to them.

SYN flooding and UDP flooding are common malicious attacks in networks. The attacks not only consume a large amount of network bandwidth and system resources of the target server but also cause network paralysis. To defend the network against the SYN flooding and UDP flooding, many new defense systems in software-defined networking (SDN) are proposed. However, most of the defense systems are only applicable to the attacks over a specific protocol such as TCP or UDP. In this paper, we therefore propose a widely applicable defense system in P4-based SDN. Through experiments, we show that the proposed defense system can effectively mitigate SYN flooding and UDP flooding. For the SYN flooding, the proposed system can release server's resources six times earlier than the related work after detecting the attack. For the UDP flooding, the proposed defense system can reduce the amount of malicious traffic by approximately two thirds compared to the related work.

Port scanning is a well-known behavior when a botnet searches target devices. To detect port scanning accurately, data with high discriminatory power are indispensable. Most related works, however, focus on data analysis methods but neglect storage limitations of switches, which makes their methods impractical. Therefore, we propose a new data collection method for collecting network information of port scanning in P4-based SDN named 0-replacement. Through simulations, we compare the 0-replacement method with two classic data collection methods. Results show that the 0-replacement method improves the true positive ratio by at least 25 percentage points but only consumes 0.36% memory space.

As the number of network devices increases rapidly, it becomes more and more difficult to defend network attacks. Large-scaled attacks, such as SYN flooding, may lead to heavy burden to the switches as well as the controller in a software defined network (SDN). In this paper, we investigate the SYN flooding and Address Resolution Protocol (ARP) spoofing attacks in SDN, and then propose mechanisms to address these two attacks. We also present a new scheme to detect SYN flooding by using only a few forwarding rules. Moreover, we utilize the Programming Protocol-independent Packet Processors (P4) technique to mitigate the burden of the controller.

Software-defined Network (SDN) is considered as a promising technology to provide a flexible network for 5G. However, as the resource of flow entries in switches is limited in SDN, a novel flow management method is required when facing massive amounts of IoT devices. To the best of our knowledge, existing methods designed for IoT can only be applied to uncomplicated network environments without multiple periodic traffic flows. In this paper, we propose Periodic Subflow-based Proactive Flow Installation Mechanism (PSPFIM), a more widely applicable method with better performance. Compared to related works, we improve the hit ratio by 45% and installation efficiency by 6.19 times according to results of simulations, which means PSPFIM can reduce the average network latency with even less flow entries.

Software-defined networking (SDN) brings new opportunities in the design of more reliable and flexible consistent network update mechanisms by making the update process independent of the underlying routing protocols. With the advance of wireless transmission technology, more and more enterprises have deployed wireless antennas into their data centers. However, to the best of our knowledge, none of the state-of-the-art solutions consider radio interference and resource competition into their model design of consistent network update. In this paper, we first propose a wireless dependency model to formulate radio interference. Based on traffic delivery and resource constraints, we formulate the resource competition problem as an update time minimization problem for ensuring the consistency properties, and then propose a greedy-heuristic scheme to minimize the total update time. Simulation results show that the proposed scheme is efficient and outperforms a well-known solution designed for network update in SDN. The simulation results also reveal the importance of resolving resource competition during network update.

In smart city, urban parking problems can be solved by sharing availability statuses of parking spaces and proper suggesting schemes. Most of the suggesting schemes recommend the nearest or the cheapest parking space from the driver's point of view. However, such a selfish decision may lead to low utilization. From the government's point of view, walking distance can be sacrificed to increase the utilization of parking spaces. In this paper, based on the open data of parking spaces in Tainan, several experiments are conducted to analyze the tradeoff between walking distance and utilization in proposed suggesting schemes.

As mobile devices and mobile applications largely grow, mobile data traffic increases at an unprecedented rate. How to alleviate mobile traffic congestion becomes a widely discussed issue. Because that most mobile data is not necessarily used immediately (e.x. update applications, download a large size video, etc.), these data are allowed to complete transmission before delay-tolerant deadline.In this paper, we propose residual size-based mobile data offloading for file transmission on mobile devices. Our method periodically checks the residual size of the file. According to the residual size and delay-tolerant deadline, our method decides how to use cellular network and WLAN(Wireless Local Area Network) for file transmission.Besides, we also propose a cost function to investigate trade-off between deadline and the fraction of the transferred size using available WLAN. We compare our method with Wifflers prediction-based offloading and On-the-spot offloading. The results show that, by the cost function, the trade-off is always worth in our method compared to Wifflers prediction-based offloading and On-the-spot offloading. Our method is more likely to guarantee file transmission completion before delay-tolerant deadline.

As the popularity of mobile applications, Wi-Fi has become one of the major access methods for many users. When a large amount of users move in public places together, such as classrooms or meeting rooms, severe load imbalance of Wi-Fi Access Points (APs) and unfair bandwidth allocation are likely to occur consequently. Software Defined Networking (SDN) is a new networking paradigm which allows the network administrators to write programs for controlling the behaviors of network devices. In this paper, we propose an adaptive load balancing scheme through association control in wireless software defined network. The proposed scheme consists of an event detection mechanism and an adaptive load balancing algorithm on controller. In our algorithm, controller can derive an optimal association solution based on the traffic load and number of users on each AP. To observe the effects of population distribution and user mobility, we propose a simulation model to investigate the performance in terms of average AP load, user bandwidth and user throughput. Our simulation results show that our scheme has better performance than other three methods and performs better in imbalanced environment.

In Long Term Evolution-Advanced (LTE-A) network, Machine Type Communications (MTC) allows machines to transfer data with almost no human intervention. In MTC, power saving is one of the most important issues. In the meantime, the 3rd Generation Partnership Project (3GPP) proposes Discontinuous Reception (DRX) mechanism to allow user equipments (UEs) to receive data only at specified time slots and turn off the radio module at other time slots. Unfortunately, current DRX mechanism is designed for normal usage of mobile users, not for MTC. In this paper, we proposes Optimistic DRX (ODRX) mechanism to allow more sleep periods for MTC devices. To allow more sleep periods, ODRX considers to release the RRC connection and re-establish the connection when the MTC device is paged. We propose analytical and simulation models for ODRX, and compare with the standard DRX through simulation experiments. The results show that, compared to standard DRX, significant extra power can be saved by sacrificing little extra wake up latency.

In Long Term Evolution-Advanced (LTE-A) network, Machine Type Communication (MTC) provides random access-based communication for automation applications. In MTC, one of the most critical issues is the emergence of large number of MTC devices which may send requests to the base station at the same time, such that severe collision occurs. When collision occurs, the message delay accumulates, and the data delivery fails. In this paper, we propose a two-stage radio access scheme to alleviate collision for MTC. MTC devices are divided into several groups, where random access in each group (as the first stage) is controlled by a device called MTC gateway. Collision of random access to the base station (as the second stage) is then effectively alleviated. To observe the effect of numbers and sizes of groups, we propose analytical model and simulation model to investigate the performance in terms of utilization of random access slots, access success probability and average message delay. Simulation results show that, by sacrificing a little message delay, the proposed scheme significantly improves access success probability.

With the explosive growth in mobile data traffic, femtocell technology is regarded as the most effective way to enhance the mobile service quality and system capacity of cellular networks. However, the major problem with femtocell deployment is finding an appropriate access control mode that mobile operators and users are willing to adopt. Among the various kinds of access control modes, the hybrid access mode is considered the most promising because it allows femtocells to give preferential access to femtocell owners, while other public users can only access femtocells with certain restrictions. Because all femtocell owners are selfish, how to provide sufficient incentives so that they will share their femtocell resources is a challenging issue. To address the problem, we propose an economic framework for mobile operator and femtocell users based on game theoretical analysis. We also exploit the concept of revenue sharing, which provides a positive cycle to sustain the femtocell service. In the framework, a femtocell game is formulated where the femtocell owners determine the proportion of femtocell resources they will share with public users, while the operator maximizes its benefit by setting the ratio of the revenue distributed to femtocell owners. We analyze the existence and uniqueness of the Nash Equilibrium of the game. The results of extensive simulations show that the proposed framework maximizes the operator's benefit and satisfies the users' service requirements.

Human mobility models strongly affect the performance in Mobile Ad Hoc Network (MANET). In order to get a more precise estimation of routing performance, it is crucial to construct a more realistic human mobility model. There are three properties investigated in literature which represent the patterns of human mobility. This paper analyzes real human traces and introduces a new property that the inter-hub movement of a human is not straight. We then construct a new random human mobility model called Irregular Orbit Levy-walk (IROL), and verify its ability to capture all of the four properties simultaneously. To investigate the feasibility of IROL, we measure routing performance of real traces, IROL, and other models in MANET on two data sets. In both data sets, IROL describes the real traces the best in terms of throughput and receive rate.

To accurately detect deadlocks in Network-on-Chip (NoC) as early as possible, a novel deadlock detection mechanism called Backward-probing Deadlock Detection (BDD) is proposed in this work, which can detect and resolve all existing deadlocks. It was realized using probe systems that generate probes for deadlock detection. A probe system includes a probe System Manager (SM) for turning on probe system, a probe Generator (GEN) for generating probes, a Link Selection (LS) connected to a Switch Allocation (SA), which is used for copying the generated probes, transmitting probes backward, and discarding probes when the probes find that the traversal path is just a congestion not a deadlock or when probe congestion occurs. There is also a TB Calculation (TBC) in LS for TB settings. Finally, a probe comparator (PB Comparator) is used for claiming deadlocks. Note that each port except the local one in a router has its own probe system.

Memory leaks generated by a running program may exhaust memory resources and degrade system performance. It is hard to reproduce these kinds of leaks, let alone to identify and fix them early in the development phase. Thus online memory leak detection is required. In this paper we propose a two-phase approach to the online memory leak detection problem for Java programs. In phase one (offline analysis), the user executes the investigated program for a while and analyzes its heap with the proposed agent. The agent then summarizes the total occupied bytes of data structures in the heap offline, and the user is then able to select classes whose object instances seem to be potential leaks by examining the total occupied bytes summary. In phase two (online monitoring), to reduce the space and time overhead at runtime, the agent monitors online the objects of selected classes instead of monitoring all objects in the heap. The approach reports the last accessed time and location (including the source file name and the line number) of each leak candidate, with which the user can identify and fix leaks in the program.

3rd Generation Partnership Project (3GPP) proposes relay architecture to extend the coverage and enhance the system capacity of Long Term Evolution Advanced (LTE-A). In the relay architecture, the header compression mechanism called RObust Header Compression (ROHC) is introduced to eliminate packet overhead and provide better spectral efficiency for IP packet transmission over the radio. However, the ROHC mechanism specified in the relay architecture causes huge computation load. This paper proposes an efficient ROHC mechanism to improve the header compression performance in the relay architecture. A modeling study is performed to tune the parameters defined in the proposed mechanism to yield better performance. Our study suggests an efficient ROHC mechanism that decreases over 30% header computation load in comparison with the mechanism defined in the specification.

As mobile technologies evolved, the mobile networks provided users not only high mobility, but also high data rates and flexible services. Meanwhile, Internet services had been very successful from 1970s to date. However, the Internet and the mobile telecommunications evolved separately and were not compatible until 3GPP proposed the IP Multimedia Subsystem (IMS). In this paper, we first introduce mobile Web 2.0 application creation in the next generation IMS network. Then we propose a proof-of-concept service called “JoinMe” on the IBM-NCTU IMS network to study how a social networking service can be created for mobile networks.

In universal mobile telecommunications system (UMTS), the core network consists of two service domains: the circuit-switched (CS) and the packet-switched (PS) domains. A call can be set up in either the CS or the PS domain. This call may be switched (transferred) between the CS and the PS domains. A major concern of domain transfer is the switching overhead. In this paper, we propose the Bearer Reservation with Preemption (BRP) scheme to support fast domain transfer, and propose a simulation model to investigate the BRP performance. Our study shows that when user behavior is irregular, the BRP scheme can effectively reduce the re-connection overhead.

The IP multimedia core network subsystem (IMS) provides multimedia services for Universal Mobile Telecommunications System (UMTS). In IMS, an emergency call is established by an emergency call session control function (ECSCF). The E-CSCF dispatches the call to the nearest public safety answering point (PSAP) according to the location of the caller. After emergency call setup, the caller's location is tracked by the PSAP through location polling. This paper investigates the performance of location tracking. Then we propose the active location reporting scheme to improve the performance of location tracking. Our study indicates that the active location reporting scheme may significantly outperform the location polling scheme.

Universal Mobile Telecommunications System (UMTS), the core network consists of two service domains: the circuit-switched (CS) and the packet-switched (PS) domains. A UMTS handset can initiate or receive a call in either the CS or the PS domain. During the call, the user may switch from one domain to another. The switching overhead is an important concern of domain transfer. In this paper, we propose the Bearer Reservation with Preemption (BRP) scheme to support fast domain transfer, and present both analytic model and simulation experiments to investigate the BRP performance. Our study indicates that when user behavior is irregular (i.e., either the variance of the domain residence times or the variance of the call holding times is large), the advantage of the BRP scheme becomes significant.

TheIP Multimedia Core Network Subsystem(IMS) provides multimedia services for Universal Mobile Telecommunications System (UMTS). In IMS, an emergency call is established by anEmergency-Call Session Control Function(ECSCF). The E-CSCF dispatches the call to the nearestPublic Safety Answering Point(PSAP) according to the location of the caller. After emergency call setup, the callers location is tracked by the PSAP through Location Polling. This paper investigates the performance of location tracking. Then we propose the Active Location Reporting Scheme to improve the performance of location tracking. Our study indicates that the Active Location Reporting Scheme may significantly outperform the Location Polling Scheme.

Push-to-talk over cellular (PoC) is a ldquowalkie- talkie-likerdquo service designed for mobile networks. In this service, the speak permission is arbitrated through the talk burst control (TBC) mechanism. This paper proposes an analytic model to study the performance of the TBC mechanism with queueing and without queueing. This analytic model is validated against simulation experiments. Through numerical examples, our study provides guidelines to set up the parameters for PoC service.

Abstract This paper proposes a client architecture for the push to talk over cellular (PoC) service based on the open mobile alliance (OMA) PoC specifications v1.0 release. We show that most standard VoIP modules can be reused for the PoC client, and the VoIP software can be easily extended to support PoC service. Then we present the detailed message flows between the PoC client and other network entities in the PoC system. A PoC client prototype has been implemented in the Industrial Technology Research Institute (ITRI) and National Chiao‐Tung University (NCTU) Joint Research Center. Copyright © 2006 John Wiley & Sons, Ltd.

Abstract-The IP multimedia core network subsystem (IMS) provides multimedia services for Universal Mobile Telecommunications System (UMTS). In IMS, any incoming call will first arrive at the interrogating call session control function (I-CSCF). The I-CSCF queries the home subscriber server (HSS) to identify the serving CSCF (S-CSCF) of the called mobile user. The S-CSCF then sets up the call to the called mobile user. This paper investigates the performance of the IMS incoming call setup. We also propose cache schemes with fault tolerance to speed up the incoming-call-setup process. Our study indicates that the I-CSCF cache can significantly reduce the incoming-call-setup delay, and checkpointing can effectively enhance the availability of I-CSCF.