With its wide applicability across a variety of enterprise verticals, the private wireless (PW) industry is growing at a rapid pace. However, existing hierarchical cellular network architectures are sub-optimal for most PW use cases due to their high computational overhead, lack of compatibility with on-premises enterprise applications, and complex management structure. To that end, this paper espouses the notion of RAN-Core Convergence for the PW user plane as an alternative to the traditional RAN-Core hierarchy through the introduction of LEVEL, a new converged approach to cellular network design. With a specific emphasis on the user plane, key highlights include a unified system architecture and protocol stack design, new primitives for interacting with the cellular control plane, and a programmable approach to QoS management. The paper also includes a high-performance systems-level prototype of the LEVEL User Plane (LEVEL-UP), deployed on an over-the-air experimental testbed. The resulting performance evaluation showcases that LEVEL-UP seamlessly handles traffic in excess of$\mathrm{\rm{20}~Gbps}$, while achieving a$60\%$reduction in compute utilization and$30\%$reduction in energy consumption when compared to state-of-the-art solutions. Furthermore, latency benchmarks reveal a delay reduction of at least$70\%$. Finally, in typical high mobility environments, LEVEL-UP helps cut bufferbloat and reduces delays from seconds to milliseconds, thereby serving as an ideal companion for the burgeoning PW industry.

With the advancement of Artificial Intelligence (AI) towards multiple modalities (language, vision, speech, etc.), multi-modal models have increasingly been used across various applications (e.g., visual question answering or image generation/captioning). Despite the success of AI as a service for multi-modal applications, it relies heavily on clouds, which are constrained by bandwidth, latency, privacy concerns, and unavailability under network or server failures. While on-device AI becomes popular, supporting multiple tasks on edge devices imposes significant resource challenges. To address this, we introduce S2M3, a split-and-share multi-modal architecture for multi-task inference on edge devices. Inspired by the general-purpose nature of multi-modal models, which are composed of multiple modules (encoder, decoder, classifier, etc.), we propose to split multi-modal models at functional-level modules; and then share common modules to reuse them across tasks, thereby reducing resource usage. To address cross-model dependency arising from module sharing, we propose a greedy module-level placement with per-request parallel routing by prioritizing compute-intensive modules. Through experiments on a testbed consisting of 14 multi-modal models across 5 tasks and 10 benchmarks, we demonstrate that S2M3 can reduce memory usage by up to 50% and 62% in single-task and multi-task settings, respectively, without sacrificing accuracy. Furthermore, S2M3 achieves optimal placement in 89 out of 95 instances (93.7%) while reducing inference latency by up to 56.9% on resource-constrained devices, compared to cloud AI.

Network slicing plays a crucial role in realizing 5G/6G advances, enabling diverse Service Level Agreement (SLA) requirements related to latency, throughput, and reliability. Since network slices are deployed end-to-end (E2E), across multiple domains including access, transport, and core networks, it is essential to efficiently decompose an E2E SLA into domain-level targets, so that each domain can provision adequate resources for the slice. However, decomposing SLAs is highly challenging due to the heterogeneity of domains, dynamic network conditions, and the fact that the SLA orchestrator is oblivious to the domain's resource optimization. In this work, we propose Odin, a Bayesian Optimization-based solution that leverages each domain's online feedback for provably-efficient SLA decomposition. Through theoretical analyses and rigorous evaluations, we demonstrate that Odin's E2E orchestrator can achieve up to 45% performance improvement in SLA satisfaction when compared with baseline solutions whilst reducing overall resource costs even in the presence of noisy feedback from the individual domains.

In recent years, the radio access network (RAN) domain has seen significant changes with increased virtualization and softwarization, driven by the Open RAN (O-RAN) movement. However, the fundamental building block of the cellular network, i.e., the base station, remains unchanged and ill-equipped to handle this architectural evolution. In particular, there exists a general lack of programmability and composability along with a protocol stack that grapples with the intricacies of the 3GPP and O-RAN specifications. Recognizing the need for an “O-RAN-native” approach to base station design, this paper introduces HexRAN– a novel base station architecture characterized by key features relating to RAN disaggregation and composability, 3GPP and O-RAN protocol integration and programmability, robust controller interactions, and customizable RAN slicing. Furthermore, the paper also includes a concrete systems-level prototype and comprehensive experimental evaluation of HexRAN on an over-the-air testbed. The results demonstrate that HexRAN uses only 8% more computing resources compared to the baseline, while managing twice the user plane traffic, delivering control plane processing latency of under 120μs, and achieving 100% processing reliability. This underscores the scalability and performance advantages of the proposed architecture.

In this paper, we study the two-level ski-rental problem, where a user needs to fulfill a sequence of demands for multiple items by choosing one of the three payment options: paying for the on-demand usage (i.e., rent), buying individual items (i.e., single purchase), and buying all the items (i.e., combo purchase). Without knowing future demands, the user aims to minimize the total cost (i.e., the sum of the rental, single purchase, and combo purchase costs) by balancing the trade-off between the expensive upfront costs (for purchase) and the potential future expenses (for rent). We first design a robust online algorithm (RDTSR) that offers a worst-case performance guarantee. While online algorithms are robust against the worst-case scenarios, they are often overly cautious and thus suffer a poor average performance in typical scenarios. On the other hand, Machine Learning (ML) algorithms typically show promising average performance in various applications but lack worst-case performance guarantees. To harness the benefits of both methods, we develop a learning-augmented algorithm (LADTSR) by integrating ML predictions into the robust online algorithm, which outperforms the robust online algorithm under accurate predictions while ensuring worst-case performance guarantees even when predictions are inaccurate. Finally, we conduct numerical experiments on both synthetic and real-world trace data to corroborate the effectiveness of our approach.

Network slicing has been widely recognized as one of the flagship use cases for Open Radio Access Network (O-RAN), enabling the provisioning of isolated network services over a shared physical infrastructure. Each slice is characterized by a set of distinct service level agreements (SLAs) tailored to meet the needs of various industries and applications. At the same time, industry-critical applications often require strict adherence to the SLA even in the worst-case scenarios. However, existing network slicing strategies merely incorporate SLA violations as penalties within the reward function, thus failing to consistently ensure perpetual SLA compliance. To address these challenges, this paper introduces PSASlicing, an intelligent resource allocation system designed for RAN slice management across the access network. More specifically, PSASlicing introduces a new reinforcement learning algorithm for maximizing resource utilization while perpetually guaranteeing the diverse SLA requirements across slices. Furthermore, PSASlicing also incorporates a trace-driven network emulator that effectively replicates the dynamic behavior of cellular networks by integrating a transition model with real-world data from an over-the-air 5G Standalone testbed. A comprehensive experimental evaluation showcases that PSASlicing achieves an average resource savings of approximately 24.0% when compared to the state-of-the-art, while guaranteeing no SLA violations.

With the open radio access network (O-RAN) movement driving the evolution of cellular networks towards 6G, real time (RT) RAN control and assurance has emerged as the next frontier in RAN programmability, with in-base station (BS) machine learning (ML) at its core. However, scalability demands associated with production-grade networks necessitate the need for a robust, yet lightweight in-BS ML operations framework to support automated ML workflows. To that end, this paper introduces TinyRIC-ML, a novel lightweight and high-performance ML platform for RT operations within the RAN. Key highlights include a comprehensive system architecture design, a concrete systems-level implementation, and a preliminary over-the-air experimental evaluation to demonstrate the system's performance and feasibility.

With the rapid deployment of indoor Wi-Fi networks, channel state information (CSI) has been used for device-free occupant activity recognition (OAR). However, various environmental factors interfere with the stable propagation of Wi-Fi signals indoors, which causes temporal variation of CSI data. In this study, we investigated temporal CSI variation in a real-world housing environment and its impact on learning-based OAR. The CSI variation over time changes distributions of the CSI data, and the pretrained model’s accuracy performance becomes degraded during long-term monitoring. In order to address the temporal dependency issue, we developed an effective long-term OAR model based on the semi-supervised meta-learning approach. Our model leveraged unlabeled target data with its pseudo labels and synthesized numerous query data sets using mixup-based data augmentation, which generalized the model during training. The model provided an average of 91.09% activity classification accuracy for the target data, which had different statistical characteristics from the source data. This result demonstrates that our model can reliably monitor occupant activities for long-term periods. The data set presented in this study is available in IEEE DataPort athttps://dx.doi.org/10.21227/z10g-vt48.

Users on edge generate deep inference requests continuously over time. Mobile/edge devices located near users can undertake the computation of inference locally for users, e.g., the embedded edge device on an autonomous vehicle. Due to limited computing resources on one mobile/edge device, it may be challenging to process the inference requests from users with high throughput. An attractive solution is to (partially) offload the computation to a remote device in the network. In this paper, we examine the existing inference execution solutions across local and remote devices and propose an adaptive scheduler, a BPS scheduler, for continuous deep inference on collaborative edge intelligence. By leveraging data parallel, neurosurgeon, reinforcement learning techniques, BPS can boost the overall inference performance by up to$8.2 \times$over the baseline schedulers. A lightweight compressor, FF, specialized in compressing intermediate output data for neurosurgeon, is proposed and integrated into the BPS scheduler. FF exploits the operating character of convolutional layers and utilizes efficient approximation algorithms. Compared to existing compression methods, FF achieves up to 86.9% lower accuracy loss and up to 83.6% lower latency overhead.

The increasing complexity of cellular networks has resulted in dynamic network performance optimization (NPO) playing a critical role in streamlining network operations. While the success of NPO techniques primarily depends upon the quality and quantity of telemetry data available from the underlying network, up until now, third-party access to such data has been largely limited due to the prevalence of proprietary interfaces throughout the access network. However, the upcoming open radio access network (RAN) architecture is set to change this trend. While a significant first step, the open RAN architecture lacks a mechanism for the generation, collection, and exposure of RAN-level statistics in a programmable manner, thereby limiting the effectiveness of dynamic NPO. To that end, this paper introduces RANSight, a system for programmable telemetry within next-generation open RANs. Key highlights include the establishment of novel primitives for programmability within RAN telemetry, a detailed component-level description of the RANSight architecture, and the design and implementation of RANSight in support of a RAN slice monitoring use case. Furthermore, the paper also includes a comprehensive experimental evaluation on an over-the-air testbed which demonstrates that not only does RANSight fulfill its stated objectives of programmable telemetry, it also significantly improves system scalability by providing a staggering 65% reduction in compute resource utilization compared to systems without RANSight.

Automating configuration is the key path to achieving zero-touch network management in ever-complicating mobile networks. Deep learning techniques show great potential to automatically learn and tackle high-dimensional networking problems. The vulnerability of deep learning to deviated input space, however, raises increasing deployment concerns under unpredictable variabilities and simulation-to-reality discrepancy in real-world networks. In this paper, we propose a novel RoNet framework to improve the robustness of neural-assisted configuration policies. We formulate the network configuration problem to maximize performance efficiency when serving diverse user applications. We design three integrated stages with novel normal training, learn-to-attack, and robust defense method for balancing the robustness and performance of policies. We evaluate RoNet via the NS-3 simulator extensively and the simulation results show that RoNet outperforms existing solutions in terms of robustness, adaptability, and scalability.

The emergence of latency-critical use cases for cellular networks has necessitated the need for real-time (RT) network performance optimization. While solutions such as the O-RAN architecture provide primitives for non-RT and near-RT control, the notion of RT control is still missing. To that end, in this paper, we introduce TinyRIC, a first-of-its-kind RT control platform for O-RAN base stations. Key highlights include a comprehensive system architecture design, a systems-level implementation in support of flexible user scheduling, and a preliminary over-the-air experimental evaluation to demonstrate the system's performance and feasibility.

Optimizing distributional utilities (such as mitigating performance tails and maximizing risk-aware objectives) is crucial for online network slice management to meet the diverse requirements of different services and applications. While Reinforcement Learning (RL) has been successfully applied to autonomous online decision-making in many network slice management problems, existing solutions often focus on maximizing the expected cumulative reward or are limited to specific distributional utilities. This paper proposes a new RL algorithm for general Distributional Utilities Optimization (DUO) in an actor-critic framework for online network slice management. In particular, we derive a DUO Temporal Difference Learning algorithm for updating distributional utilities in the critic through stochastic gradient descent. It is proven that the Distributional Optimal Bellman Operator for distributional utilities is a γ-contraction and thus is guaranteed to converge. In addition, we parameterize the policy by another neural network and prove a revised policy gradient theorem for distributional utilities, which shows that the derived policy update converges to at least a stationary point of the DUO problem. Our proposed algorithm works with arbitrary smooth utility functions on the return distributions, making it suitable for optimizing various network slice performance objectives in an online setting. Our solution is implemented and validated by building a hybrid trace-driven network simulator, which was built using an open-source O-RAN dataset, along with data collected from a 5G O-RAN testbed. Results demonstrate a significant improvement over heuristic and RL baselines.

Network slicing achieves cost-efficient slice customization to support heterogeneous applications and services. Configuring cross-domain resources to end-to-end slices based on service-level agreements, however, is challenging, due to the complicated underlying correlations and the simulation-to-reality discrepancy between simulators and real networks. In this paper, we propose Atlas, an online network slicing system, which automates the service configuration of slices via safe and sample-efficient learn-to-configure approaches in three interrelated stages. First, we design a learning-based simulator to reduce the sim-to-real discrepancy, which is accomplished by a new parameter searching method based on Bayesian optimization. Second, we offline train the policy in the augmented simulator via a novel offline algorithm with a Bayesian neural network and parallel Thompson sampling. Third, we online learn the policy in real networks with a novel online algorithm with safe exploration and Gaussian process regression. We implement Atlas on an end-to-end network prototype based on OpenAirInterface RAN, OpenDayLight SDN transport, OpenAir-CN core network, and Docker-based edge server. Experimental results show that, compared to state-of-the-art solutions, Atlas achieves 63.9% and 85.7% regret reduction on resource usage and slice quality of experience during the online learning stage, respectively.

The increasing popularity of programmable wireless networks has led to efforts by both academia as well as industry to redesign access networks based on the disaggregated Open Radio Access Network (O-RAN) concept. However, the absence of a development platform for prototyping O-RAN use cases and the perceived complexity of the O-RAN specification have led to a stagnation of systems research efforts in this domain with a disproportionate focus on monolithic RAN architectures. With a view to overcoming these challenges, this paper introduces the ProSLICE platform, a full-scale realization of the complete O-RAN specification based on purely open source components along with several enhancements and extensions for fine-grained network control and ease of use. Key contributions include a fully disaggregated O-RAN-compliant RAN, a custom O-RAN service model in support of network slicing, statistics and configuration applications for the O-RAN RAN Intelligent Controller (RIC), and a comprehensive use case-based performance evaluation.

While Open RAN harbors the potential to revolutionize cellular access networks, the absence of an open platform for use case prototyping and a near-exclusive focus on 5G SA have tempered its commercial appeal. With a view to addressing these limitations, this paper introduces a disaggregated O-RAN platform with a novel multi-RAT RAN slicing framework across LTE, 5G NSA, and 5G SA. A preliminary performance evaluation showcases promising results addressing key features related to end-user service quality and use case-specific adaptability.

Network slicing allows mobile network operators to virtualize infrastructures and provide customized slices for supporting various use cases with heterogeneous requirements. Online deep reinforcement learning (DRL) has shown promising potential in solving network problems and eliminating the simulation-to-reality discrepancy. Optimizing cross-domain resources with online DRL is, however, challenging, as the random exploration of DRL violates the service level agreement (SLA) of slices and resource constraints of infrastructures. In this paper, we propose OnSlicing, an online end-to-end network slicing system, to achieve minimal resource usage while satisfying slices' SLA. OnSlicing allows individualized learning for each slice and maintains its SLA by using a novel constraint-aware policy update method and proactive baseline switching mechanism. OnSlicing complies with resource constraints of infrastructures by using a unique design of action modification in slices and parameter coordination in infrastructures. OnSlicing further mitigates the poor performance of online learning during the early learning stage by offline imitating a rule-based solution. Besides, we design four new domain managers to enable dynamic resource configuration in radio access, transport, core, and edge networks, respectively, at a timescale of subseconds. We implement OnSlicing on an end-to-end slicing testbed designed based on OpenAirInterface with both 4G LTE and 5G NR, OpenDayLight SDN platform, and OpenAir-CN core network. The experimental results show that OnSlicing achieves 61.3% usage reduction as compared to the rule-based solution and maintains nearly zero violation (0.06%) throughout the online learning phase. As online learning is converged, OnSlicing reduces 12.5% usage without any violations as compared to the state-of-the-art online DRL solution.

Network slicing is a promising technology that allows mobile network operators to efficiently serve various emerging use cases in 5G. It is challenging to optimize the utilization of network infrastructures while guaranteeing the performance of network slices according to service level agreements (SLAs). To solve this problem, we propose SafeSlicing that introduces a new constraint-aware deep reinforcement learning (CaDRL) algorithm to learn the optimal resource orchestration policy within two steps, i.e., offline training in a simulated environment and online learning with the real network system. On optimizing the resource orchestration, we incorporate the constraints on the statistical performance of slices in the reward function using Lagrangian multipliers, and solve the Lagrangian relaxed problem via a policy network. To satisfy the constraints on the system capacity, we design a constraint network to map the latent actions generated from the policy network to the orchestration actions such that the total resources allocated to network slices do not exceed the system capacity. We prototype SafeSlicing on an end-to-end testbed developed by using OpenAirInterface LTE, OpenDayLight-based SDN, and CUDA GPU computing platform. The experimental results show that SafeSlicing reduces more than 20% resource usage while meeting SLAs of network slices as compared with other solutions.

Network slicing was adopted as a solution for future networks to support various applications with diverse requirements. While active research has focused on this functionality, most of the work targets RAN or packet core slicing and leaves the transport network nearly untouched. With packet core functions moving to data centers and parts of RAN functions moving to edge clouds, the transport network will gain significantly more importance. To ensure stringent service level agreements (SLAs) and facilitate slice management, it is imperative for the transport network to support live slice mobility with no disruption of current services. In this paper, we present FestNet, a Flexible and Efficient Sliced Transport Network, achieved by our virtualized Programmable Data Plane (vPDP) and two-layer design. Not only conventional protocols but also stateless and stateful network functions (NFs) can be integrated as slices in FestNet. We implemented a FestNet prototype. Our evaluation shows that FestNet supports live slice migration with no packet loss and no state loss for stateful slices and that FestNet provides many times faster slice operations than implementations in related work.

5G technologies promise to revolutionize mobile networks and push them to the limits of resource utilization. Besides better capacity, we also need better resource management via virtualization. End-to-end network slicing not only involves the core but also the Radio Access Network (RAN) which makes this a challenging problem. This is because multiple alternative radio access technologies exist (e. g. ,LTE, WLAN, and WiMAX), and there is no unifying abstraction to compare and compose from diverse technologies. In addition, existing work assumes that all RAN infrastructure exists under a single administrative domain. Software-Defined Radio Access Network (SD-RAN) offers programmability that facilitates a unified abstraction for resource sharing and composition across multiple providers harnessing different technology stacks. In this paper we propose a new architecture for heterogeneous RAN slicing across multiple providers. A central component in our architecture is a service orchestrator that interacts with multiple network providers and service providers to negotiate resource allocations that are jointly optimal. We propose a double auction mechanism that captures the interaction among selfish parties and guarantees convergence to optimal social welfare in finite time. We then demonstrate the feasibility of our proposed system by using open source SD-RAN systems such as EmPOWER (WiFi) and FlexRAN (LTE).

5G is set to revolutionize the network service industry with unprecedented use-cases in industrial automation, augmented reality, virtual reality and many other domains. Network slicing is a key enabler to realize this concept, and comes with various SLA requirements in terms of latency, throughput, and reliability. Network slicing is typically performed in an end-to-end (e2e) manner across multiple domains, for example, in mobile networks, a slice can span access, transport and core networks. Thus, if an SLA requirement is specified for e2e services, we need to ensure that the total SLA budget is appropriately proportioned to each participating domain in an adaptive manner. Such an SLA decomposition can be extremely useful for network service operators as they can plan accordingly for actual deployment. In this paper we design and implement an SLA decomposition planner for network slicing using supervised machine learning algorithms. Traditional optimization based approaches cannot deal with the dynamic nature of such services. We design machine learning models for SLA decomposition, based on random forest, gradient boosting and neural network. We then evaluate each class of algorithms in terms of accuracy, sample complexity, and model explainability. Our experiments reveal that, in terms of these three requirements, the gradient boosting and neural network algorithms for SLA decomposition out-perform random forest algorithms, given emulated data sets.

Fog computing is a paradigm which brings computing, storage, and networking closer to end users and end devices for better service provisioning. One of the crucial factors in the success of fog computing is on how to incentivize the individual users' edge resources and provide them to end users such that fog computing is economically beneficial to all involved economic players. In this paper, we model and analyze a market of fog computing, from which we aim at drawing practical implications to uncover how the fog computing market should operate. To this end, we conduct an economic analysis of such user-oriented fog computing by modeling a market consisting of Infrastructure and Service Provider (ISP), end Service Users (SUs), and Edge Resource Owners (EROs) as a non-cooperative game. In this market, ISP, which provides a platform for fog computing, behaves as a mediator or a broker which leases EROs' edge resources and provides various services to SUs. In our model, a two-stage dynamic game is used where in each stage, there exists a dynamic game, one for between ISP and EROs and another for between ISP and SUs, to model the market more practically. Despite this complex game structure, we provide a closed-form equilibrium analysis which gives an insight on how much economic benefit is obtained by ISP, SUs, and EROs from user-oriented fog computing under what conditions, and we figure out the economic factors that have a significant impact on the success of fog computing.

We consider the problem of optimally mapping an edge computing service that is modeled as a tree with multiple processing sub-tasks and data flows onto the underlying physical network. As new computing and data analytics applications require more complicated data processing structures, and different types of data (e.g., images, videos, and numbers) sensed at geographically distributed locations must be collected and processed to obtain a complex and comprehensive result, highly intelligent algorithms are needed to solve this challenging problem. In this paper, we propose a learning-based hierarchical service tree placement strategy that aims to optimize the net utility, defined as achieved utility minus network congestion. The key idea is to decouple a service tree into appropriate sub-trees each containing a single computing sub-task as well as associated data flows and to recursively leverage Q-learning to place each sub-tree while maintaining the dependencies of sub-tasks in the service tree structure. It enables a scalable solution for large networks with unknown arrival statistics and complex service structures. Numerical results show that our solution can significantly outperform baseline heuristics in online service tree placement.

Softwarization enables tremendous flexibility for networks as the use of software-defined networking (SDN) and programmable data planes (e.g. P4) together support dynamic reconfiguration of networks in real-time, in response to network conditions and new service requests. In such networks, it is imperative to ensure that end-to-end network services continue to satisfy SLAs, especially performance requirements. However, in the presence of unpredictable dynamics due to network reconfigurations, it is not possible to guarantee prior to deployment that a service will meet such SLAs. Run-time verification - in combination with programmable control and data plane monitoring - can provide a basis for detecting potential performance SLA violations, together with identifying and executing appropriate network mitigations. In this paper, we propose a verification transverse based on formal specifications, that spans performance SLAs across the distributed SDN control and programmable data planes, and can coordinate with both planes to execute dynamic reconfiguration that mitigate the detected issues. We demonstrate a proof-of-concept prototype based on an extension of the Aerial run-time verification tool, together with the Inband Network Telemetry (INT) capability, on a network running distributed ONOS controllers together with a P4 data plane.

Fog computing is a paradigm which brings computing, storage, and networking closer to end users and devices for better service provisioning. One of the crucial factors towards the success of fog computing is how to incentivize the individual users' edge resources, thereby opening the era of user- participated fog computing. In this paper, we provide an economic analysis of such user-oriented fog computing by modeling a market consisting of ISP (Infrastructure and Service Provider), SUs (end Service Users), and EROs (Edge Resource Owners) as a noncooperative game. In this market, ISP, which provides a platform of fog computing, behaves as a mediator or a broker to lease the edge resources from EROs and provide various services to SUs. In our game formulation, a two-stage dynamic game is used, where in each stage there exists another dynamic game, one for between ISP and EROs and another for between ISP and SUs, to model the market more practically. Despite this complex game structure, we provide a closed- form equilibrium analysis, which gives an insight of how much economic benefits are obtained by ISP, SUs, and EROs under what conditions.

In future mobile networks, e.g., 5G, emerging IoT services are expected to support billions of IoT devices with unique characteristics and traffic patterns. In this paper we propose an SDN-based IoT Mobile Edge Cloud Architecture (SIMECA1) which can deploy diverse IoT services at the mobile edge by leveraging distributed, lightweight control and data planes optimized for IoT communications. We prototyped our architecture using a pre-commercial mobile networking software stack to demonstrate the feasibility and utility of our approach.

Named Data Networking (NDN) uses content names as routing entries, and thus the scalability of NDN routing is of primary concern. NDN allows in-network caching as a built-in functionality; however, if network nodes make caching decisions individually, duplicate copies of the same content may exist among nearby nodes. To address these problems, we propose Coordinated Routing and Caching (CoRC) that mitigates routing scalability and enhances the efficiency of the in-network storage. CoRC aligns the routing and caching mechanisms to manage the same content namespace for better performance. We evaluate CoRC (and its variants) with Vanilla NDN in terms of the cache hit ratio, hop count, and traffic load by running software routers on Amazon EC2. To demonstrate the feasibility of CoRC, we also implement and test the processing time of CoRC forwarding in Linux machines.

Consider a network of prosumers of media content in which users dynamically create and request content objects. The request process is governed by the objects' popularity and varies across network regions and over time. In order to meet user requests, content objects can be stored and transported over the network, characterized by the capacity and energy efficiency of the storage and transport resources. The energy efficient dynamic in-network caching problem aims at finding the evolution of the network configuration, in terms of the content objects being cached and transported over each network element at any given time, that meets user requests, satisfies network resource capacities and minimizes overall energy use. We provide 1) an information-centric optimization framework for the energy efficient dynamic in-network caching problem, 2) an offline solution, EE-OFD, based on an integer linear program (ILP) that obtains the maximum efficiency gains that can be achieved with global knowledge of user requests and network resources, and 3) an efficient fully distributed online solution, EEOND, that allows network nodes to make local caching decisions based on their current estimate of the global energy benefit. Our solutions take into account the network heterogeneity, in terms of capacity, energy efficiency and content popularity, and adapt to changing network conditions minimizing overall energy use.

In content-centric networking (CCN), the in-network caching feature provides several attractive advantages such as low dissemination latency and network transport load reduction. Thus, CCN requires less transport energy but additional energy to provide a caching capability at every content router. In this paper, we investigate the minimum energy consumption that CCN can achieve with optimal cache locations by considering different caching hardware technologies, number of downloads per hour, and content popularity. We first set up an energy consumption model for CCN and then formulate linear and nonlinear programming problems to minimize total energy consumption of CCN. Also, a genetic algorithm (GA) approach is proposed to find energy-efficient cache locations. Using reported energy efficiency of computational hardware and network equipment, we show CCN yield greater energy savings for very popular content and small-sized catalog, compared to conventional CDN. Our results also indicate that two aspects of the memory technology, energy-proportional caching and sufficient memory capacity, are critical to the overall energy efficiency gain of CCN.

In recent years, the data center networks commonly accommodate applications such as MapReduce and web search that inherently shows the incast communication pattern; multiple workers simultaneously transmit TCP data to a single aggregator. In this environment, the TCP performance is significantly degraded in terms of goodput and query completion time, as a result of the severe packet loss at Top of Rack (ToR) switches. The TCP senders aggressively transmit packets causing throughput collapse even though the network pipe size, i.e., bandwidth-delay product, is extremely small. In this paper, we introduce a novel end-to-end congestion control algorithm called IA-TCP that avoids the TCP incast congestion problem effectively. IA-TCP employs the rate-based algorithm at the aggregator node, which controls both the window size of workers and ACK delay. Through extensive NS-2 simulations, we validate that our algorithm is scalable in terms of the number of workers achieving enhanced goodput and zero timeouts.

Surveillance systems are developed to enhance security and safety by constantly observing locations of interest. Although those systems can observe scenes from each camera separately, it is difficult to keep track of any moving target across different cameras. This paper firstly proposes Video Diary Service (VDS) to solve this problem. VDS is an automatic diary service, which makes it possible to keep track of users' lives. In addition, VDS can identify social networking relationships among the users while each camera is watching multiple users. By exploiting these properties of VDS, we extend VDS into a new surveillance system called S-VDS. We also illustrate a few application scenarios where the proposed system can enhance security and safety.

In wireless local area network (WLAN)-based vehicular networks, the performance anomaly problem is serious due to random channel access among different vehicles with diverse channel conditions and association times. To address the performance anomaly problem, we propose a signal to noise ratio (SNR)-based admission control scheme where only vehicles with better channel conditions (or higher transmission rates) are serviced by access points (APs) and thus the impact of vehicles with low transmission rates can be mitigated. The starvation issue of rejected vehicles can be resolved by considering mobility in vehicular environments with multiple intersections. Simulation results demonstrate that the SNR-based admission control scheme can improve the network throughput and the starvation problem diminishes as the number of intersections increases users.

With the advent of various multimedia streaming applications requiring reliability and high bandwidth, multicasting in wireless LANs has been gaining more attentions as the supported bit rate increases. However, according to the specification of the IEEE 802.11 standard, broadcast/multicast frames are transmitted at a fixed and low bit rate due to the absence of a feedback mechanism such as ACK. This simple broadcasting technique with no feedback signal raises some issues; reliability, efficiency and fairness. In this paper, hence, we propose a framework for multicasting termed Leader-Based Multicast Service (LBMS) to alleviate those limitations. LBMS consists of a leader-based transmission and feedback mechanism for multicasting by extending the IEEE 802.11v standard, which is the next-generation standard for network management. Also, we try to support legacy 802.11 stations can still participate in multicasting. Simulation exhibits that the gain of the proposed multicasting scheme increases as (i) more stations compete for the channel, and (ii) the wireless channel condition becomes poorer.

Multicasting is useful for various applications such as multimedia broadcasting. In current 802.11, multicast frames are sent as broadcast frames at a low transmission rate without any acknowledgement or binary exponential backoff. This naive multicasting mechanism degrades the performance of not only multicast flows but also unicast flows. In this paper, we propose a new multicasting mechanism based on the leader- based approach to improve the legacy multicast transmissions, maintaining coexistence with legacy 802.11 devices. Simulations show that our protocol achieves well-balanced performance in terms of reliability, latency, goodput, and transmission fairness in comprehensive environments.

Finding a reliable and efficient routing path in vehicular ad hoc networks (VANETs) is a challenging issue due to high mobility of vehicles and frequent link breakage. Motivated by this, we propose a robust and efficient routing protocol, called MMR. The contribution of this paper is two-fold: two-level routing and a new routing metric. A routing process of MMR consists of the macro level and the micro level. MMR forwards a packet to an approximate location of the destination at the macro level and then forwards a packet to the exact location of the destination at the micro level. This two-level routing reduces the protocol overhead and improves scalability in terms of the number of nodes. MMR also introduces a new routing metric that reduces the protocol overhead and path breakage by considering velocities of vehicles. Through simulations, we show that MMR improves the routing performance by about 30~40% in highly mobile environments, compared to the existing ad hoc routing protocols such as AODV and GPSR.

The IEEE 802.11e standard supports a direct-link setup (DLS) mechanism optionally to improve the throughput. Using this mechanism, IEEE 802.11e stations in proximity can directly exchange frames with no intervention of an access point. However, extensive simulations reveal the severe unfairness between external and local TCP connections in the IEEE 802.11e DLS mode due to the characteristics of TCP: a window-based flow control mechanism. This paper first analyzes why a fairness problem happens between external and local TCP connections in the IEEE 802.11e DLS mode. Then, we seeks to achieve fairness between them by introducing a novel mechanism dubbed half direct-link setup plus (H-DLS) which differentiates the paths for TCP DATA and ACK packets of local TCP connections. Simulation results reveal that H-DLS achieves the fairness between external and local TCP connections while keeping the aggregate throughput higher than the original IEEE 802.11 infrastructure.

This paper via analysis and simulation revisits the interaction between MAC contention and TCP congestion control over IEEE 802.11 WLANs, misled in the previous efforts. The results reveal that the effective number of contending wireless stations is not proportional to the number of wireless stations with an upstream TCP flow in IEEE 802.11 wireless LANs. Thus, we propose a new scheme called TCP ACK priority (TAP) in which, by allowing an access point to transmit TCP ACKs at the highest priority, the optimal number of competing stations are allowed to contend for media access to utilize link bandwidth efficiently. We use an ns-2 simulator to evaluate the performance of TAP with the IEEE 802.11 DCF. The results show that there is an improvement in network performance without the loss of fairness between upstream TCP flows. The extensions for IEEE 802.11e/n are also considered.

Abstract — The IEEE 802.11 protocol has become the de facto standard in wireless LANs. However, it reveals the unfairness problem between unicast flows and multicast flows since multicast packets are not subject to binary exponential backoff. To prevent a multicast flow from overwhelming the wireless link bandwidth is a crucial issue. This paper seeks to achieve fairness between unicast and multicast flows by introducing Unicast-Friendly Multicast (UFM). The central idea behind the UFM algorithm is to dynamically change the contention window size for multicast packets to limit the bandwidth share of a multicast flow equal to that of a unicast flow. The proposed UFM algorithm adjusts the multicast contention window size depending on the number of competing stations. We present two versions of UFM: the first one calculates the multicast window size by inferring the average contention window size of unicast flows, while the second one maintains the mapping table between the number of competing stations and the corresponding multicast window size given by related analysis. Simulation reveals that both versions of UFM achieve the fairness by providing almost the fair share of bandwidth to each flow regardless of unicast or multicast under the saturated network conditions. I.

International audience

This paper introduces a new split-TCP approach for improving TCP performance over IEEE 802.11-based wireless LANs. TCP over wireless LANs is not aggressive, which is a fundamental reason for poor performance. We propose augmented split-TCP (AS-TCP) to mitigate this problem. Our scheme extends the split-connection approach that divides a connection into two different connections at a split point such as an access point. Using AS-TCP, a mobile host emulates TCP ACK packets using MAC ACK frames, instead of receiving real TCP ACK packets. We compared AS-TCP with both normal TCP and I-TCP (indirect TCP) by simulation. Results show that AS-TCP achieves higher throughput, fairer resource allocation and, in power-saving mode, shorter delays.

IEEE 802.11 MAC protocol in distributed coordination function (DCF) mode is not scalable as the number of contending users increases. In this paper, we propose two new contention resolution schemes to reduce the number of collisions due to a large number of contending users. The first scheme, hybrid DCF (H-DCF), splits the monolithic contention resolution into two phases to enhance scalability. The other scheme, enhanced hybrid DCF (EH-DCF), provides fairness with the current IEEE 802.11 MAC and yet improves the efficiency of contention resolution. The major contribution of our work is to provide inter-operability with the IEEE 802.11 DCF as well as an efficient channel access, unlike the previous works. We conduct comprehensive simulation experiments by using an ns-2 simulator to demonstrate the advantages of these two schemes. The results show that they improve network performance by up to 10~30%, compared to the original IEEE 802.11 DCF. Therefore, H-DCF and EH-DCF can be good candidates for MAC protocols for hotspots where numerous users access wireless LANs intensively

To join an IP network and communicate with others, a node needs to be configured either manually by an administrator or automatically through a DHCP server. However, the former method is impractical for large networks, while the latter is infeasible in the case of a mobile ad hoc network due to the mobility of the nodes. This paper introduces two distributed IP address auto-configuration mechanisms for mobile ad hoc networks, namely (a) RADA (random address allocation) and (b) LiA (linear allocation). RADA is based on random IP address selection, while LiA linearly assigns new addresses by utilizing the current maximum IP address value. We have also introduced an improved version of LiA, known as LiACR (linear allocation with collision resolution), which reduces control overhead. Then, we discuss extensions of these mechanisms capable of handling network partitioning and merging. Performance evaluations of RADA, LiA and LiACR were conducted through simulation. The results related to address allocation time and control overhead are presented and compared.

The integration of 3G networks and WLAN as complementary has begun to attract much attention in industry as well as academia. This topic is becoming a burning issue, and one of the key questions which it raises is how to support a seamless vertical handoff. This paper introduces a new interface selection algorithm for energy-efficient vertical handoff in tightly coupled systems capable of supporting seamless handoff. Our proposed scheme, Wise Interface Selection (WISE) switches the active network interface, after taking into consideration the characteristics of the network interface cards and the current level of data traffic, with the cooperation of the mobile terminals and network. Interface switching operates independently on both the downlink and the uplink for the purpose of energy conservation. We show through simulation that less energy is consumed with WISE than when only a 3G network or WLAN interface is used, resulting in a longer lifetime for the mobile terminals. In the case of TCP connections, additional throughput gain can also be obtained.

In this paper, we propose a novel technique for reducing the energy consumption in idle mode. Our proposed technique called the network capacity adaptation derives an effective network capacity required by a given application and logically slows down the network capacity when the traffic load is low. By slowing down the logical network capacity, the neighboring mobile stations can stay longer in the power-off mode, significantly saving the energy consumed in idle mode.