Vehicular Technology and OpenStack NFV A Word On The Edge Cloud - - PowerPoint PPT Presentation

Vehicular Technology and OpenStack NFV

A Word On “The Edge Cloud”

PRAKASH RAMCHANDRAN GOKHALE JAYANTHI

01.11.2017

What is VT?

Vehicular Technology (VT) addresses part of Intelligent Transportation System (ITS)

What are the VT use cases?

Near-term is Connected Vehicles and long term is Autonomous Driving and in-between some Entertainment & I-IOT applications

What is Openstack’s Edge Stack

Openstack Edge Stack using Small & Medium footprints

How VT uses NFV?

VT fits into using NFV MANO to manage Edge Stack

Using OpenStack Edge for CVA

How do you get OpenStack?

Conclusion & Q&A’s

Next Steps- Need Scoping & Blueprints for VT Support at Edge

Agenda

What is VT?

Vehicular Technology involves :

• Communication Radar, Lidar, V2X, (Advanced

driver-assistance systems)ADAS, Autonomous Vehicles (AV) , Short Range Data Communication (SRDC) standards

• Cloud systems and devices to collect and respond

with precise calculations - latency limitations to respond for steering and driving decisions.

• Location, Maps & Routes Maps and traffic patterns

and various smart transportation Data to store and analyze in Openstack Cloud using interface and work with different radio technologies.

• Openstack existing and evolving edge standards

with strategies to enable local and global decisions along with its IoT, Sensors and efforts to support devices meeting LTE Release 12-14 PC5 & LTE-V standards.

Congestion Cost in B$, $121 , 15% lost lives B$, $75 , 9%

Vacation Loss B$, $500 , 63%

Freight Cost B$, $100 , 13%

Potential Cost to save using Autonomous Vehicles R&D in USA

Intelligent Transportation System and VT

Vehicular Technology is part of transportation and logistic enablement for next generation

Sensors Summary

Ultrasonic Sensor: low resolution, short distance for parking assist RADAR Sensor: radio wave detection and ranging. Medium resolution to determine obstacle, their distance, their speed Odometer Sensor: to measure the speed of the vehicle Prebuilt Maps: containing road / terrain information Inertial Navigation System: uses gyroscope and accelerometers to determine vehicle’s position, orientation and speed. Works along with GPS. DSRC: Direct Short Range Communication, used for vehicle to vehicle(V2V) and Vehicle to Infrastructure (V2I) communication [road, traffic light etc]. Cameras: Grabbing pictures & video for determining the scene IR Sensor: Used to capture low intensity images in the dark / night GPS: Global Positioning System to determine vehicle’s position (lat – long) LIDAR: Light Detection and ranging. Uses light / laser beam based detection of obstacles and prepares a 3D contour map.

Autonomous Vehicles

➡

Vehicular Ad-hoc Networks(VANET)

• As vehicles fall out of the signal range and drop out of

the network, other vehicles can join in, connecting vehicles to one another so that a mobile Internet is created.

• A Vehicular Ad-Hoc Network, or VANET is a

technology that uses moving vehicles as nodes in a network to create a mobile network. VANET turns every participating vehicle into a wireless router or node, allowing vehicles approximately 100 to 300 meters of each other to connect and, in turn, create a network with a wide range.

• VANET is a subgroup of MANET where the nodes

refer to vehicles.

VANET Environment

• RN

CN DN1 DN-Central

Principal Components

• OBU – Onboard units (for onboard interactions and

V2V communication by messages sent directly or through RSU)

• RSU – Road Side Units (for V2V and V2I). Receives

messages from vehicles and broadcasts them to other vehicles or the Infrastructure

• Internet Gateway Infrastructure

Dedicated Short Range Communication (DSRC)

• 802.11p Add wireless access to vehicular networks

and implements OSI stack

• Wireless Protocol with Licensed band of 5.9GHz, 7

channels, Range of 1000m, Data rate 6 to 27Mbps

• Mainly used in communication of
• 1) Vehicle to Vehicle
• 2) Vehicle to Roadside Infrastructure

Communication Types at a Glance

VEHICLE TO VEHICLE COMMUNICATION (V2V)

• V2V is most suited for short range vehicular networks using DSRC.
• It is Fast and Reliable and provides real time safety
• It does not need any roadside Infrastructure.
• V2V does not have the problem of Vehicle Shadowing in which a smaller

vehicle is shadowed by a larger vehicle preventing it to communicate with the Roadside infrastructure.

• Location based broadcast and multicast are the proper communication

methods for collision avoidance in V2V Communication

• Issues:
• Stringent response times (Maneuvering Time)
• Identification / Addressing of vehicles (use Certificate of Authority and

assign an address tagged the number plate)

(V2I) +ROADSIDE COMMUNICATION (V2R)

• Vehicle to Infrastructure provides solution to longer-range vehicular networks.
• It makes use of preexisting network infrastructure such as wireless access

points (Road-Side Units, RSUs).

• Communications between vehicles and RSUs are supported by Vehicle-to-

Infrastructure (V2I) protocol and Vehicle-to-Roadside (V2R) protocol.

• Needs large area coverage and more feature enhancements for Vehicle

Applications using DSRC or other technologies (Bluetooth, WIFI etc.).

• Roadside infrastructure involves establishing a connectivity to a

communication infrastructure (5G / 4G or otherwise) and hence involves additional installation costs.

Security in DSRC

• Scheduler in every OBU
• Pre-emptive policy for Higher Priority

Messages

• Each OBU has a valid certificate issued

by CA(Certification Authority) based on unique license plate registration

• Based on Digital signature sent by

OBU, using Public key decryption message is verified

Routing Methodologies

• In V2V communication, the collision warning messages are broadcast from

vehicle to vehicle across multiple hops without the involvement of a roadside unit.

• In case of V2R the warning messages are first sent to a roadside unit, and then

broadcast by the roadside unit to all vehicles in range.

• In V2R/V2V Hybrid Model, Vehicles which receive a warning message via V2V

communication will send it to a roadside unit if they did not receive a warning message with the same event ID from roadside units.

@OpenStack

Connected Vehicle Applications (CVA)

Next?

• pen stack
• penstack

OpenStackFoundation

Cloud Infrastructure for Intelligent Vehicles

• OBU – On Board Unit
• RSU – Road Side Unit
• RSUC – Road Side Unit Controller
• Edge Cloud
• AU – Application Unit
• Connectivity V2I, V2V, I2V, I2I

OBU OBU OBU OBU OBU OBU RSU RSU RSU RSU RSU BS BS

RSUC RSUC Orchestration & Network Management

V2V I2I V2V V2I V2I V2I V2I I2I RSU I2I

Road Side Unit Controller (RSCU)

• Acts like a local data centre, with virtualized resources.
• Cloudlet infrastructure
• Runs multiple AUs in VMs
• Deploys Aus to the RSUs on demand
• Containerised instances for AUs
• Hence needs VN, VM, VS (Nova/Neutron/Cinder/Ceph/Swift)

RSU Deployment

• Dense Regions: At every intersection. Vehicles with dashboard antenna

range is approximately 50 metres.

• Normal Regions: 200 meters with an outdoor antenna. Hence RSU can be

deployed every 200 m

• Sparse Regions: Every 1/2 mile

V2X Connected Vehicle Applications CVA) Low latency, Location Based, Moderate throughput

• V2I Safety
• Red Light Violation Warning
• Curve Speed Warning
• Stop Sign Gap Assist
• Spot Weather Impact Warning
• Reduced Speed/Work Zone Warning
• Pedestrian in Signalized
• Crosswalk Warning (Transit)
• V2V Safety
• Emergency Electronic Brake Lights (EEBL)
• Forward Collision Warning (FCW)
• Intersection Movement Assist (IMA)
• Left Turn Assist (LTA)
• Blind Spot/Lane Change Warning (BSW/LCW)
• Do Not Pass Warning (DNPW)
• Vehicle Turning Right in Front of Bus Warning

(Transit)

20

http://www.its.dot.gov/pilots

@OpenStack

Openstack Modules and Resources for “OPEN EDGE STACK”

Next?

• penstack
• penstack

OpenStackFoundation

NFV Infrastructure & Services at the Edge

Mobility & Session Management

Policy Management for

Service Offerings and associated resource requirements (OPA + Congress) MEC/ Access/Metro Ethernet connectivity Central Large Data Centers(Public & Private) IP Multimedia Services (IMS)

User Plane Functions SFC/Classifiers

Distributed Openstack (MPP) Cloud Management

systems Edge Enabled Applications Transportation Management System and Location Based Safety Applications

(Services/AF)

MANO NFVO Orchestration (ONAP, OpenBaton, OSM, K8, OPNFV CI/Testing)

Middleware APIs

Cloudlet, Statelet, Netlet

“OPEN EDGE STACK”

built from Infrastructure Microservices Distributed “OPEN EDGE NFVI” with VNF/Containers Openstack Lightweight/

Medium Weight VIM

Configurations

In scope of Openstack In Scope for Upstream In Scope common to all

Openstack Module Mapping: AU & support it needs

Typical AU function & features

• Application Unit:
• Deploy applications for local operations, VANET

signalling for VANET messages to other vehicles (V2V)

• Most of these applications are connected to
• nboard sensors.
• Applications for safety interact with RSU for inputs

like (road blocks, lane restrictions, accident, etc). These applications should be dynamically deployed.

• AU can be implemented as multiple AFs for Safety,

Traffic Analytics, Security/firewall & load balancer, packet classifiers to classify flows & transfer traffic from road side to the centralized Unit(CU) through core (offload service))

• RSU needs multiple service instances, hence

requires VN, VM, VS (Nova/Neutron/Cinder/Ceph/Swift)

Continue.

OpenStack Components & Emerging Ones

• Keystone for ID management /
• service catalog management (will be at controller)
• Glance (global)
• Distributed nova at the RSU
• Distributed neutron at RSU
• Distributed cinder / ceph / swift at RSU

Emerging Microservices Proposals For DC - Cloudlet For State Management - Statelet For Edge Network - Netlet Evolving For Orchestration/ZUN, Kuryr, Magnum Evolving For Image Management

• Kola

Cloud Infrastructure for Intelligent Vehicles

• OBU – On Board Unit
• RSU – Road Side Unit
• RSUC – Road Side Unit Controller
• Edge Cloud
• AU – Application Unit
• Connectivity V2I, V2V, I2V, I2I

OBU OBU OBU OBU OBU OBU RSU RSU RSU RSU RSU BS BS

RSUC RSUC Orchestration & Network Management

V2V I2I V2V V2I V2I V2I V2I I2I RSU I2I

• VT can be supported by NFV based Open Edge Stack
• Questions?

@OpenStack

THANKS.

Questions?

• penstack
• penstack

OpenStackFoundation

Specification for Edge Cloud Infrastructure

Typical RSU/ RSU-C(2U-20U) specs (Medium)

• RSU infrastructure:
• CPU: multi core 64 bit CPU capable of multi threading @ 2 GHz [LS2045 –

LS2085 having 8 Cortex A-57 CPUs NXP processors]

• Memory: 8 GB RAM
• Storage: 128 GB SDRAM
• N/W:
• Downlink – DSRC or WIFI (multilink infrastructure with RFID, BT, WIFI,

DSRC)

• Uplink – WiMax / WIFI to a access point connection to internet

Typical OBU specs (Small)

• OBU infrastructure:
• CPU : dual core CPU capable of multi threading @ 512 MB
• Memory: 2-4 GB RAM
• Storage: 16 GB SDRAM
• Sensors interconnect – Direct coupling / CAN
• Networking:
• Downlink: WIFI for connection to mobiles
• Uplink: WIFI, DSRC, multilink connection (BT, RFID, DSRC, WIFI)

using 802.11p WAVE