The Tactile Internet presently constitutes a vision of an Internet over which, in addition to current communications modalities, a sense of touch can be transported. In that case, people would no longer need to be physically near the systems they operate, but could control them remotely. The main problem that needs to be solved to realize the Tactile Internet is summarized by the “1 ms challenge.” If the response time of a system is below 1 ms, the end-user will not be able to tell the difference between controlling a system locally or from another location. This paper offers a summary of the requirements for haptic communications, followed by an overview of challenges in realizing the Tactile Internet. In addition, possible solutions to these challenges are proposed and discussed. For example, the development of the fifth generation mobile communication networks will provide a good foundation upon which a Tactile Internet could be built. This paper also describes the design of a modular testbed needed for testing of a wide variety of haptic system applications.

/pdf/challenges-in-haptic-communications-over-the-tactile-1996sbtfn8.pdf

Challenges in Haptic Communications Over the Tactile Internet

The fifth generation of cellular networks (5G) will rely on edge cloud deployments to satisfy the ultra-low latency demand of future applications. In this paper, we argue that such deployments can also be used to enable advanced data-driven and Machine Learning (ML) applications in mobile networks. We propose an edge-controller-based architecture for cellular networks and evaluate its performance with real data from hundreds of base stations of a major U.S. operator. In this regard, we will provide insights on how to dynamically cluster and associate base stations and controllers, according to the global mobility patterns of the users. Then, we will describe how the controllers can be used to run ML algorithms to predict the number of users in each base station, and a use case in which these predictions are exploited by a higher-layer application to route vehicular traffic according to network Key Performance Indicators (KPIs). We show that the prediction accuracy improves when based on machine learning algorithms that rely on the controllers’ view and, consequently, on the spatial correlation introduced by the user mobility, with respect to when the prediction is based only on the local data of each single base station.

Machine Learning at the Edge: A Data-Driven Architecture With Applications to 5G Cellular Networks

The Internet has made several giant leaps over the years, from a fixed to a mobile Internet, then to the Internet of Things, and now to a Tactile Internet. The Tactile Internet goes far beyond data, audio and video delivery over fixed and mobile networks, and even beyond allowing communication and collaboration among things. It is expected to enable haptic communications and allow skill set delivery over networks. Some examples of potential applications are tele-surgery, vehicle fleets, augmented reality and industrial process automation. Several papers already cover many of the Tactile Internet-related concepts and technologies, such as haptic codecs, applications, and supporting technologies. However, none of them offers a comprehensive survey of the Tactile Internet, including its architectures and algorithms. Furthermore, none of them provides a systematic and critical review of the existing solutions. To address these lacunae, we provide a comprehensive survey of the architectures and algorithms proposed to date for the Tactile Internet. In addition, we critically review them using a well-defined set of requirements and discuss some of the lessons learned as well as the most promising research directions.

https://hal.archives-ouvertes.fr/hal-02963827/document

A Comprehensive Survey of the Tactile Internet: State-of-the-Art and Research Directions

Fulcrum coding combines a high-field outer Random Linear Network Coding (RLNC) that generates outer coding expansion packets with a small-field inner RLNC that combines the source packets and the outer coding expansion packets. This two-layer Fulcrum coding allows flexible decoding in receivers with heterogeneous computational capabilities. Fulcrum coding has so far only been studied for conventional dense RLNC, which randomly selects all coding coefficients, and only for a statically fixed number of outer expansion packets. However, the probability that the coding coefficient row of a newly received packet is linearly independent of prior received coding coefficient rows (a prerequisite for successful decoding) is highly dynamic. We propose to exploit the dynamics of this probability to reduce the computational complexity of Fulcrum coding. In particular, we vary the density of non-zero coding coefficients, i.e., equivalently, the sparsity of coding coefficients, and the number of outer expansion packets to keep the complexity low while maintaining a reasonably high decoding probability. We introduce the general principles of dynamic sparsity and expansion packets (DSEP) for Fulcrum coding as well as two specific example DSEP policies. Our evaluations indicate that DSEP Fulcrum can increase the encoding throughput tenfold and increase the decoding throughput 1.4 to 4.3 fold while achieving decoding probabilities that are typically less than 1% lower than the conventional Fulcrum decoding probabilities. We also find that DSEP achieves somewhat higher encoding and decoding throughputs than the CodornicesRq (Release 2.1) implementation of RaptorQ block coding for small blocks (generations) of source packets, while RaptorQ is substantially faster for large generation sizes. Furthermore, we develop and evaluate an elementary DSEP recoding mechanism that achieves a recoding throughput more than double the decoding throughput.

/pdf/dsep-fulcrum-dynamic-sparsity-and-expansion-packets-for-duybg2egur.pdf

DSEP Fulcrum: Dynamic Sparsity and Expansion Packets for Fulcrum Network Coding

Future applications such as driverless cars, industrial Internet and smart grids will demand high bandwidth, resilience, security and low latency communication at the same time. Those technical requirements will be met by the 5G communication system, which is not only focusing on the future air interface but also the deployment in the core network. The paradigm shift from agnostic store-and-forward towards intelligent networks paves the ground for the mobile edge cloud — a concept to not only place cloud computing in close proximity to the users or things, such as in the base stations or even in access points in the houses, but also move the cloud with the user as the device moves along the network. This would enable minimum latency services, which require advanced migration techniques that facilitate not just quick but very frequent migrations. In this paper, we study and compare the state of the art migration techniques offered by virtualization tools like docker and KVM and propose an application level migration protocol that eliminates the drawbacks of the former. We also present the implementation of the proposed protocol in a latency sensitive gaming application, where the server is migrated live during the gameplay between hosts, transparently to the users, as a proof of concept and study the handover in detail.

On the study and deployment of mobile edge cloud for tactile Internet using a 5G gaming application

This paper is advocating for joint design of com- munication and control in cyber physical systems, especially self driving automobiles, exploiting adaptive and next generation networking concepts such as mobile edge clouds in 5G commu- nication systems. In this paper, we present our '5G connected- cars' testbed that was demonstrated in multiple international events including the Mobile World Congress, Barcelona in 2016. It constitutes a complete network architecture and cloud software services that facilitates custom built miniature cars to run autonomously on a track with an intersection without collisions by using hybrid car-to-car and car-to-infrastructure communi- cation. We tackle the latency problem in classical connected cyber physical systems by the use of mobile edge cloud in the full blown emulated 5G air interface in the cmWave band in Nokia Airframe. Finally, presenting some measurements from the testbed, we advocate for a dedicated 5G air interface for reliable and scalable steering and control, as opposed to the use of ISM bands such as IEEE802.11P.

Joint Design of Communication and Control for Connected Cars in 5G Communication Systems

The Multi-access Edge Computing (MEC) paradigm allows several automotive applications to be offloaded from the vehicles to the edge. Besides a higher computation capability, compared to the on-board vehicle, and the shorter latency, compared to the remote cloud, the edge offers additional (context) information that is not directly available at the vehicle, e.g., via data fusion from multiple sources. In this paper we propose a high-level architecture for MEC-assisted platooning control. Within the architecture, the longitudinal controller is conceived as a virtualized application running on an edge server, and aligned with the European Telecommunications Standard Institute (ETSI) MEC reference framework. Performance assessment conducted through a realistic simulation framework, coupling a vehicular mobility simulator and Docker containers, showcases the feasibility and effectiveness of our proposal.

A MEC-assisted Vehicle Platooning Control through Docker Containers

Vehicle platooning is one of the challenging communication system due to the mobility of vehicles and wireless channels. Therefore, the channel quality of communication is changing quickly, as a result providing high resilient and low latency vehicle-to-vehicle (V2V) communication is still an open research topic. In order to address the problem, we did a measurement with two transporters driven on a public high way. The distance was between 100-300 m and the speed was in the range of 100-150 km/h. We used IEEE802.11p standard and measured packet reception ratio (PRR) of different data rates. Moreover, we implement systematic sliding window random linear network coding (RLNC) in intermediate nodes in order to transmit packets in platooning for low delay and high resilience. Systematic sliding window RLNC is so far applied only to end-to-end communication. The delay and loss performance of this coding scheme has not been evaluated on multi-hop communication. In this work, we use the measured PRR to mimic the real channel performance as well as apply sliding window coding scheme for a 6 vehicle platooning scenario. Our simulation results show that it is possible to keep the erasure amount below to 15% with 85% probability and overall delay performance below the total transmission times with 75% probability until 4th following vehicle with a constant data rate of 6 MBit/s.

Vehicle Platooning: Sliding Window RLNC for Low Latency and High Resilience.

Fulcrum codes, a variant of random linear network coding (RLNC), especially designed for heterogeneous environment. A Fulcrum encoder combines two codes operating on both small, such as binary and big, such as 28 Galois fields. That allows the Fulcrum decoder to operate on either the smaller or greater field size, depending on whether it has a moderate or stronger computing power. However, setting on one particular Galois field is inefficient for end devices, especially in broadcasting scenarios when feedback to senders is normally impractical. We propose an adaptive decoding process that alters among Galois fields to decode. Our comprehensive evaluation shows that the proposed adaptive decoding significantly reduces the computation complexity at end devices while simultaneously maintaining a high decoding probability.

Adaptive Decoding for Fulcrum Codes

Recently private 5G networks so called 5G campus networks have gained a lot of public attention. Typically, such networks are associated to be implemented in huge production companies to control assembly lines. However, we present a demonstrator for the deployment of a private 5G network in a SME. Our demonstrator was designed in cooperation with a local bakery to relocate dough pieces in the middle of the night. This is an example of a monotones task, which alikes can be found in even the smallest companies. Such tasks can be better solved by robots, providing human workers more time for varied and creative tasks. We also point out some options how SMEs can get 5G campus networks at reasonable costs.

Christopher Lehmann

Papers

Joint Design of Communication and Control for Connected Cars in 5G Communication Systems

A MEC-assisted Vehicle Platooning Control through Docker Containers

Vehicle Platooning: Sliding Window RLNC for Low Latency and High Resilience.

Adaptive Decoding for Fulcrum Codes

BakeryRobot: 5G connected robot for SMEs