We apply a fluid-reservoir model to study the Age-of-Information (AoI) of update packets through energy-harvesting transmitters. The model is closer to how energy is stored and depleted in reality, and can reveal the system behavior for different settings of packet arrival rates, service rates, and energy charging and depletion rates. We present detailed results for both finite and infinite transmitter buffers and an infinite energy reservoir, and some indicative results for a finite reservoir. The results are derived for the mean AoI in the case of an infinite transmitter buffer and an infinite reservoir, and for the mean peak AoI for the remaining cases. The results show that, similar to a system without energy constraints, the transmitter buffer should be kept to a minimum in order to avoid queueing delays and maintain freshness of updates. Furthermore, a high update packet rate is only helpful in energy-rich regimes, whereas in energy-poor regimes more frequent updates deplete the energy reservoir and result in higher AoI values.
5G New Radio (NR) represents a shift in mobile telephony whereby the network architecture runs containerized software on commodity hardware. In preparation, numerous 4G software stacks have been developed to test the containerization of core network functions and the interfaces with radio access network (RAN) protocols. In this work one such stack, developed by the OpenAirInterface Software Alliance (OSA), is used to create a low-cost, simplified mobile network. Commercial off-the-shelf (COTS) user equipment (UE) is then connected to the network to demonstrate how a major buffer overflow vulnerability present in certain Global Navigation Satellite System (GNSS) chipsets can be leveraged to enable a spoofed network attack. Finally, the theoretical attack method is extended to 5G NR networks.
The last few years have seen an explosive growth in the Internet of Things (IoT) as billions of new low-powered devices are connected to the network in every sector, from household items to sensors, healthcare devices, and industrial controls. Devices often rely on vertical architectures, with each type of device requiring a separate platform for control, data sharing, and storage. New horizontal architectures are needed to allow sharing of resources between devices to optimize hardware efficiency. In addition, new networking paradigms such as Software Defined Networking (SDN) are required to efficiently manage increased delay-sensitive network demand. This broad range of new technologies and skills make entry difficult for students and new researchers, and innovative practical testbeds for IoT systems and SDN will be required for training and research. We propose an IoT testbed with multiple networking layers and heterogeneous devices to simultaneously support networking research, anomaly detection, and security principles applied specifically to IoT for education and research, providing a complex yet practical hands-on environment made entirely of open-source tools and Commercial Off-The-Shelf (COTS) materials that can be replicated for use by others seeking to build such a system.
In this paper, we propose a software defined network (SDN) architecture suitable for optical intra-rack data center networks (DCNs). Thus in an optical DCN, we distinguish the intra-rack communication from the inter-rack one and we focus on the intra-rack network This is implemented by the passive optical interconnection among the rack servers using a set of dedicated data wavelengths. In this way, low power consumption is required. A SDN controller is introduced to play an administrative role in the optical intra-rack network, by determining the servers' transmission coordination. Core of the proposed SDN configuration is the optical communication link that is assumed to connect each rack server with the SDN controller, aiming to ensure the perfect synchronization among the control and the data planes operations. Based on this assurance, we propose a synchronous transmission SDN-based pre-transmission coordination medium access control (SPC-MAC) protocol that the SDN controller follows in order to instruct the servers to transmit without packet losses due to collisions in the intra-rack network In this way, high throughput and low delay are achieved for diverse intra-rack network configurations. Simulation results prove that the proposed software defined optical intra-rack DCN in conjunction with the proposed SPC-MAC protocol reach high bandwidth exploitation, while they provide high scalability, access fairness and reliability.
Users are often interested in a specific type of data (user-preferred data) from large volumes of data streaming into the system in real-time. An efficient system that only stores user-preferred data from these data can reduce storage space by storing only the user-preferred data and discarding the remaining data. It also reduces the search latency which allows the users to search for relevant information in a timely manner. The motivation behind this research is to devise a technique that filters streaming data and stores only the filtered data for post-processing, thereby saving storage space and reduce search latency. A proof-of-concept prototype for this technique has been built on Apache Spark by leveraging the parallel processing framework for streaming data and its machine learning library. The performance of the prototype subjected to synthetically generated streaming data is analyzed. The analysis of experimental results demonstrates the efficacy of this technique and provides insights into system behavior and performance.
Mobility management is a key function of modern wireless networks. Specifically, in vehicular environments which are characterized by high mobility, networks must be able to provide a stable, uninterrupted connection to the vehicles to guarantee the continuity of their services. In this paper, the Proxy Mobile IPv6 (PMIPv6) as well as its enhanced schemes suggested by the research literature are studied, while at the same time the architecture of the 5G Core (5GC) network is described. Subsequently, a novel handover (HO) scheme is proposed for supporting Vehicle to Infrastructure (V2I) communications. The proposed scheme is based on the Fast Proxy Mobile IPv6 (FPMIPv6) protocol using the functionalities of the 5GC network and applies signaling enhancements and transient binding to improve the route of data packets to the mobile node (MN), aiming at the optimization of both Predictive and Reactive HOs. Evaluation results shows that the proposed scheme outperforms the FMIPv6 protocol in terms of HO delay, HO cost and packet loss.
As the distributed computing paradigm continues to evolve due to advances such as cloud and fog computing, the need for effective workload allocation on distributed resources has become more imperative than ever. However, workload orchestration in distributed environments poses many challenges, especially in the case where the workload comprises jobs consisting of parallel component tasks. A particularly challenging scenario in the context of such bag-of-tasks applications, is the case where a component task produces an output that does not meet a predefined Quality of Service (QoS) threshold. In such a case, a new task is dynamically spawned at runtime, in order to carry out the additional computational work required for producing an acceptable result. In order to leverage data locality, the dynamically spawned task must be assigned to the same resource as the previous task that failed to meet the QoS requirements. Consequently, it is important to examine how a such dynamically changing workload affects the performance and load balancing of the distributed resources. To this end, in this paper we investigate the orchestration of bag-of-tasks applications with dynamically spawned tasks in a distributed environment. Three task routing techniques are studied and compared using simulation, under various load cases.
With the gradually popular high-speed wireless networks and 5G environments, the quality and reliability of network services will be suited for mobile vehicles. In addition to communicating information between vehicles, they can also communicate information with surrounding roadside equipment, pedestrians or traffic signs, and thus improve the road safety of passers-by. Recently, various countries have continuously invested in research on autonomous driving and unmanned vehicles. The open communication environment of the Internet of Vehicles in 5G will expose all personal information in the field of wireless networks. This research is based on the consideration of information security and personal data protection. We will focus on how to protect the real-time transmission of information between mobile vehicles to prevent from imbedding or altering important transmission information by unauthorized vehicles, drivers or passers-by participating in communications. Moreover, this research proposes a multi-level security key management agreement based on a dynamic M-tree structure for Internet of Vehicles to achieve flexible and scalable key management on large-scale Internet of Vehicles.
This paper presents the concept of sharing a hyper-visor address space with a standard Linux program. In this work, we add hypervisor awareness to the Linux kernel and execute code in the HYP exception level through using the hyplet. The hyplet is an innovative way to code interrupt service routines and remote procedure calls under ARM. The hyplet provides high performance and run-time predictability. We demonstrate the hyplet implementation using the C programming language on an ARM8v-a platform and under the Linux kernel. We then provide performance measurements, use cases, and security scenarios.
New network architectures like 5G adopt new service models where more providers share the infrastructure and offer different network services for users. The slicing concept in 5G networks is important to provide flexible, scalable, and on-demand solutions for the vast array of applications in networks. In this paper, we present a multilayer network model that allows the Service Function Chaining (SFC) with the support of multi-tenant slices. To analyze the Quality of Service in the slices, we set up a network with few virtual routers over Amazon Web Services (AWS) and use a variety of technologies such as Segment Routing (SR) for implementing Service Chains. We analyze SFC algorithms with different capabilities in avoiding overloads on the network links. The analysis shows the impact of traffic load increase on packet losses and the impact of deploying more slices on the same infrastructure.
Recent technological advances, such as cloud and fog computing, have made prevalent the utilization of distributed computational resources. There is a variety of complex workloads that need to be processed on such platforms. The type of the workload may vary over time. Two of the most important challenges in a distributed environment are efficient resource allocation and the employment of appropriate workload scheduling techniques. In this paper, we examine dynamic scheduling of complex workloads on distributed resources. The workload consists of single-task jobs, bag-of-task jobs and linear workflows. The characteristics of the investigated workload vary over time. Two scheduling techniques are studied. Simulation modeling is used to compare their performance under different workload and system load scenarios.
This paper presents a four-element multiple-input multiple-output (MIMO) antenna with slotted ground plane and patch for 5G millimeter waves communication systems. It is designed using Computer Simulation Technology (CST) software on Rogers RT/Duriod 5880 with a thickness and dielectric constant of 0.787mm and 2.2, respectively. The inset feeding technique is applied to obtain the matching impedance of 50Ω. The overall size of the antenna is 48×12×0.787mm3, and the dimensions of the slots on the ground plane and patch are determined through applying the empirical Hill Climbing algorithm to achieve an enhanced bandwidth and gain. It resonates at 26.455GHz with a reflection coefficient of -65.09dB and provides a huge bandwidth of 4.468GHz in the range of 23.37GHz – 27.838GHz at S11 ≤ -10dB. It also achieves a mutual coupling of less than -32dB without utilizing a decoupling structure or extra spacing. In addition, other MIMO antenna parameters such as Envelope Correlation Coefficient (ECC), Diversity-Gain (DG), and Channel Capacity Loss (CCL) are evaluated which satisfy 5G systems requirements.
Storage system traces are important for examining real-world applications, studying potential bottlenecks, as well as driving benchmarks in the evaluation of new system designs. While file system traces have been well-studied in earlier work, it has been some time since the last examination of the SMB network file system. The purpose of this work is to continue previous SMB studies to better understand the use of the protocol in a real-world production system in use at the University of Connecticut. The main contribution of our work is the exploration of I/O behavior in modern file system workloads as well as new examinations of the inter-arrival times and run times for I/O events. We further investigate if the recent standard models for traffic remain accurate. Our findings reveal interesting data relating to the number of read and write events. We notice that the number of read and write events is significantly less than creates and the average number of bytes exchanged per I/O is much smaller than what has been seen in previous studies. Furthermore, we find an increase in the use of metadata for overall network communication that can be taken advantage of through the use of smart storage devices.
Indoor localization has received attention due to the emergence of location-based mobile applications. In addition to mobile application, indoor localization can be used for other purposes, such as optimization of building operations. We present new models to study the performance of a localization method based on the Received Signal Strength Indicator (RSSI), using Long Term Evaluation-Advanced (LTE-A) Ultra Dense Networks (UDNs). Our approach provides a localization technique with existing infrastructure. Fingerprinting is used to estimate the location of User Equipment (UE) in a building from their RSSI value. The simulation results show that LTE-A UDNs can provide good accuracy for indoor location.
Cognitive Autonomous Networks (CAN) advance network automation by using Cognitive Functions (CFs) which learn optimal behavior through interaction with the network. However, as in self Organizing Networks (SON), CFs encounter conflicts due to overlap in parameters or objectives. Owing to the nondeterministic behavior of CFs, their conflicts cannot be resolved using SON-style rule-based approaches. This paper proposes the Cognitive Bargaining Mechanism (CBM) as the optimal generic way for resolving - any type of conflict among CFs, conflict among any number of CFs and any number of simultaneously existing conflicts among CFs. With the CAN modeled as a multi-agent system (MAS), CBM uses Nash’s Social Welfare Function (NSWF) to compute a compromise among CFs that is fair and optimal for the collective interest of the system. To prove the feasibility of the approach, we model three different CAN scenarios in Python and show the resulting configurations when a CBM-enabled controller is used to resolve all the possible conflicts in the CAN.
In the field of cost optimization in cloud computing infrastructure, different strategies are used to calculate Virtual Machine (VM) allocations to support workloads while minimizing costs. Usually, VM allocations strategies are focused on determining the number and types of VMs required to support workloads at every moment, but generally they lack procedures to account for the storage costs of VMs. A significant part of these costs is generated by the storage of the Virtual Machine Images (VMIs) required to deploy VMs. In this paper, we present an improvement to a state-of-the-art VM allocation strategy, by integrating VMI costs. To achieve this, the allocation model of the strategy is extended. Then a wide set of experiments is carried out to study the improvements in the cost optimization. Several factors are analyzed and the most significant ones identified. The experimentation shows savings up to 20%, but are very dependent on the VMI sizes, the characteristics of the workload and the cloud infrastructure.
This paper proposes a three-dimensional (3D) communication channel model for an indoor environment considering the effect of the Hypersurface. The Hypersurface is a software controlled intelligent metasurface, which can be used to manipulate electromagnetic waves, as for example for nonspecular reflection and full absorption. Thus it can control the impinging rays from a transmitter towards a receiver location in both LOS and NLOS paths, e.g. to combat distance and improve wireless connectivity. We focus on the 60 GHz mmWave frequency band due to its increasing significance in 5G/6G networks and evaluate the effect of Hypersurface in an indoor environment in terms of attenuation coefficients related to the Hypersurface reflection and absorption functionalities, using CST simulation, a 3D electromagnetic simulator of high frequency components. To highlight the benefits of Hypersurface coated walls versus plain walls, we use the derived Hypersurface 3D channel model and a custom 3D ray-tracing simulator for plain walls considering a typical indoor scenario for different Tx-Rx location and separation distances.
In this paper, we consider a multi-pair two-way full-duplex relaying system with multiple-input-multiple-output (MIMO) users. Each pair of users exchange information with the aid of a massive MIMO amplify-and-forward (AF) relay, and correlation between the antennas both at the users as well as the relay is considered. The direct link between all user nodes is assumed to be non-negligible, which is suitable for practical urban scenarios. The low-complexity transceiver design at the relay based on maximum ratio combining/maximum ratio transmission (MRC[MRT) processing is presented. The performance of the system is evaluated in terms of the achievable sum-spectral efficiency under two communication schemes. The first scheme attempts to make use of the joint benefits of the relayed and direct links, while in the second scheme the direct link is considered as interference. Comparison analysis show that the first scheme outperforms the second in the presence of a strong direct link. Moreover, the detrimental effect of spatial correlation between the antennas at each user node is investigated.
In recent years, improvements in high-speed Analog-to-Digital Converters (ADC) and sensor technology has encouraged researchers to improve the performance of Data Acquisition (DAQ) systems for scientific experiments which require high speed and continuous data measurements — in particular, measuring the electronic and magnetic properties of materials using pump-probe experiments at high repetition rates. Experiments at TELBE are capable of acquiring almost 100 Gigabytes of raw data every ten minutes. The DAQ system used at TELBE partitions the raw data into various subdirectories for further parallel processing utilizing the multicore structure of modern CPUs. Furthermore, several other types of processors that accelerate data processing like the GPU and FPGA have emerged to solve the challenges of processing the massive amount of raw data. However, the memory and network bottlenecks become a significant challenge in big data processing, and new scalable programming techniques are needed to solve these challenges. In this contribution, we will outline the design and implementation of our practical software approach for efficient parallel processing of our large data sets at the TELBE user facility.
In order to optimize costs in cloud computing deployments, Virtual Machine (VM) allocation strategies are frequently employed. To compare different strategies, workload traces are required. These traces should be of one second resolution (to take into account the per-second billing) and of one year length (to consider reserved VMs properly). However, there are no public traces with these characteristics corresponding to transactional services. To overcome this issue, in this paper we present different synthesis techniques designed to generate a trace of one second resolution based on a trace with lower resolution, usually one hour. The influence of the different synthesis techniques in the cost optimization process is analyzed, concluding that they have an appreciable effect in the optimization process. As an alternative, the execution of VM allocation strategies using traces shorter than a year (a few months, for example) and the extrapolation of the results to the whole year is also analyzed. We found that this mode of operation can generate misleading results.
