Internet-Draft | Vehicular Mobility Management | August 2023 |
Jeong, et al. | Expires 8 February 2024 | [Page] |
This document specifies a Vehicular Mobility Management (VMM) scheme for IP-based vehicular networks. The VMM scheme takes advantage of a vehicular link model based on a multi-link subnet. With a vehicle's mobility information (e.g., position, speed, acceleration/deceleration, and direction) and navigation path (i.e., trajectory), it can provide a moving vehicle with proactive and seamless handoff along with its trajectory.¶
This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79.¶
Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet-Drafts is at https://datatracker.ietf.org/drafts/current/.¶
Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress."¶
This Internet-Draft will expire on 8 February 2024.¶
Copyright (c) 2023 IETF Trust and the persons identified as the document authors. All rights reserved.¶
This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (https://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Revised BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Revised BSD License.¶
This document proposes a mobility management scheme for IP-based vehicular networks, called Vehicular Mobility Management (VMM). The VMM is tailored for a vehicular network architecture and a vehicular link model described in the IPWAVE problem statement document [RFC9365].¶
Vehicle Neighbor Discovery (VND) is proposed as Extended IPv6 Neighbor Discovery (ND) for IP-based vehicle networks [ID-IPWAVE-VND] to support the vehicle-to-vehicle or the vehicle to Road-Side Unit (RSU) interactions. For an efficient IPv6 Stateless Address Autoconfiguration (SLAAC) [RFC4862], VND adopts an optimized Address Registration using a multihop Duplicate Address Detection (DAD). This multihop DAD enables a vehicle to have a unique IP address in a multi-link subnet consisting of multiple wireless subnets with the same IP prefix, which corresponds to wireless coverage of multiple RSUs. VND also supports IP packet routing over a connected Vehicular Ad Hoc Network (VANET) by allowing vehicles to exchange the prefixes of their internal networks through their external wireless interface.¶
The mobility management in this multi-link subnet needs a new approach from the legacy mobility management schemes. This document aims at an efficient mobility management scheme called VMM to support efficient V2V, V2I, and V2X communications in a road network. The VMM takes advantage of the mobility information (e.g.,a vehicle's speed, direction, and position) and trajectory (i.e., navigation path) of each vehicle registered in the Traffic Control Center (TCC) of the vehicular cloud.¶
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in RFC 2119 [RFC2119].¶
This document uses the terminology described in [RFC4861] and [RFC4862]. In addition, the following new terms are defined as below:¶
This section describes a vehicular network architecture for V2V and V2I communication. A vehicle and an RSU have their internal networks including in-vehicle devices or servers, respectively.¶
A vehicular network architecture for V2I and V2V is illustrated in Figure 1. In this figure, there is a vehicular cloud containing a TCC. The TCC has Mobility Anchors (MAs) responsible for the vehicles mobility management. Each MA is in charge of the mobility management of vehicles under its prefix domain, which is a multi-link subnet of RSUs sharing the same prefix [RFC9365]. A vehicular network is a wireless network consisting of RSUs and vehicles. The RSUs are interconnected via a wired network, allowing vehicles to build VANETs via V2V and V2I communications.¶
In Figure 1, three RSUs are deployed either at intersections or along roadways. They are connected to an MA through wired networks. The vehicular network has two subnets, such as Subnet1 and Subnet2. Subnet1 is a multi-link subnet consisting of multiple wireless coverage areas of multiple RSUs, which share the same IPv6 prefix to construct a single logical subnet [RFC9365]. That is, the RSU1 and RSU2 wireless links belong to Subnet1. Thus, since Vehicle2 and Vehicle3 use the same prefix for Subnet1, and that are within the wireless communication range, they can communicate directly with each other. Note that in a multi-link subnet, a vehicle (e.g., Vehicle2 and Vehicle3 in Figure 1) can configure its global IPv6 address through an address registration procedure that includes the multihop DAD specified in VND [ID-IPWAVE-VND].¶
Subnet2 on the other hand, uses a different prefix than Subnet1. Vehicle4 residing in Subnet2 cannot communicate directly to Vehicle3 because it belongs to a different subnet. Vehicles can construct a connected VANET so they can communicate with each other without relaying on RSU, but on the forwarding over the VANET. In the case where two vehicles belong to the same multi-link subnet, but they are not connected in the same VANET, they can use RSUs. In Figure 1, even though Vehicle1 is disconnected from Vehicle3, they can communicate indirectly with each other through RSUs such as RSU1 and RSU2.¶
This document specifies a mobility management scheme for the vehicular network architecture, as shown in Figure 1. Vehicle2 is supposed to communicate with the corresponding node denoted as CN1, and Vehicle2 is moving in the wireless coverage of RSU1. When Vehicle2 moves out of the coverage of RSU1 and moves into the coverage of RSU2 where RSU1 and RSU2 share the same prefix, packets sent by CN1 should be routed through RSU2 to Vehicle2. Also, when Vehicle2 moves out of the coverage of RSU2 and moves into the coverage of RSU3 where RSU2 and RSU3 use two different prefixes, the CN1 packets should be delivered to Vehicle2 via RSU3. A handoff procedure allows a sender's packets to be delivered to a destination vehicle which is moving within the wireless coverage areas.¶
This section explains the detailed procedure of mobility management of a vehicle in a road network as shown in Figure 1.¶
A mobility management is required for the seamless communication of vehicles moving between the RSUs. When a vehicle moves into the coverage of another RSU, a different IP address is assigned to the vehicle, the transport-layersession information (i.e., an end-point's IP address) is reconfigured to avoid service disruption. Considering this issue, this document proposes a handoff mechanism for seamless communication.¶
In [VIP-WAVE], the authors constructed a network-based mobility management scheme using Proxy Mobile IPv6 (PMIPv6) [RFC5213], which is highly suitable for vehicular networks. This document uses a mobility management procedure similar to PMIPv6, but uses a newly proposed Shared-Prefix model in which vehicles in the same subnet share the same prefix.¶
Figure 2 shows the binding update flow when a vehicle entered the RSU subnet. The RSUs act as Mobility Anchor Gateway (MAG) defined in [VIP-WAVE]. When it receives an RS message from a vehicle containing its mobility information (e.g., position, speed, and direction), an RSU sends a Proxy Binding Update (PBU) message to its MA [RFC5213][RFC3775]. This contains a Mobility Option including the vehicle's mobility information. The MA receives the PBU and sets up a Binding Cache Entry (BCE) as well as a bi-directional tunnel (denoted as Bi-Dir Tunnel in Figure 2) between the serving RSU and itself. Through this tunnel, all traffic packets to the vehicle are encapsulated toward the RSU. Simultaneously, the MA sends back a Proxy Binding Acknowledgment (PBA) message to the serving RSU. This serving RSU receives the PBA and sets up a bi-directional tunnel with the MA. After this binding update, the RSU sends back an RA message to the vehicle. This message includes the RSU's prefix for the address autoconfiguration of the vehicle.¶
When the vehicle receives the RA message, it performs the address registration procedure including a multihop DAD for its global IP address based on the prefix announced by the RA message according to the VND [ID-IPWAVE-VND].¶
In PMIPv6, an MA (i.e., LMA) allocates a unique prefix to each vehicle to guarantee the uniqueness of each address, but in this document, an MA allocates in its domain a unique IP address to each vehicle with the same prefix through the multihop-DAD-based address registration. This unique IP address allocation ensures that vehicles own unique IP addresses in a multi-link subnet and can reduce the waste of IP prefixes in legacy PMIPv6.¶
When the vehicle changes its location and its current RSU (denoted as c-RSU) detects that the vehicle is moving out of its coverage, the c-RSU reports the leaving of the vehicle to the MA and de-register the binding via PBU.¶
With this report, the MA will send back a PBA to notice the de-register to c-RSU, and get ready to detect new binding requests. If MA can figure out the new RSU (denoted as n-RSU) based on the vehicle's trajectory, it will directly change the end-point of the tunnel into n-RSU's IP address for the vehicle.¶
Figure 3 shows the handoff of a vehicle within one prefix domain (i.e., a multi-link subnet) with PMIPv6. As shown in the figure, when the MA receives a new PBU from the n-RSU, it changes the tunnel's end-point from the c-RSU to n-RSU. If there are ongoing IP packets toward the vehicle, the MA encapsulates the packets and then forwards them towards n-RSU. Through this network-based mobility management, the vehicle is not aware of any changes at its network layer and can maintain its transport-layer sessions without any disruption.¶
If c-RSU and n-RSU are adjacent, that is, vehicles are moving in specified routes with fixed RSU allocation, the procedure can be simplified by constructing the bidirectional tunnel directly between them (cancel the intervention of MA) to alleviate the traffic flow in MA as well as reduce handoff delay.¶
Figure 4 shows a vehicle handoff within one prefix domain (as a multi-link subnet) with DMM [RFC8885]. The RSUs are in charge of detecting when a node joins or moves to its domain. If the c-RSU detects that the vehicle is going to leave its coverage and to enter the area of an adjacent RSU, it sends a PBU message to inform n-RSU of the vehicle's handoff. If n-RSU receives the PBU message, it constructs a bidirectional tunnel between c-RSU and itself, and then sends back a PBA message as an acknowledgment to c-RSU. If there are ongoing IP packets toward the vehicle, c-RSU encapsulates the packets and then forwards them to n-RSU. When n-RSU detects the entrance of the vehicle, it directly sends an RA message to the vehicle so that the vehicle can assure that it is still connected to a router with its current prefix. If the vehicle sends an RS message to n-RSU, n-RSU responds to the RS message by replying to the vehicle with an RA .¶
When a vehicle moves from a prefix domain to another prefix domain, a handoff between multiple prefix domains is required. As shown in Figure 1, when Vehicle3 moves from the subnet of RSU2 (i.e., Subnet1) to the subnet of RSU3 (i.e., Subnet2), a multiple domain handoff is performed through the cooperation of RSU2, RSU3, MA1 and MA2.¶
Figure 5 shows the handoff of a vehicle between two prefix domains (i.e., two multi-link subnets) with PMIPv6. When the vehicle moves out of its c-RSU belonging to Subnet1, and moves into the n-RSU belonging to Subnet2, c-RSU detects the vehicle's leaving and reports it to MA1. MA1 figures out that the vehicle will get into the coverage of the n-RSU based on its trajectory and sends MA2 a PBU message to inform MA2 that the vehicle will enter the coverage of n-RSU belonging to MA2. MA2 sends a PBA message to n-RSU to inform that the vehicle will enter the coverage of n-RSU along with handoff context such as c-RSU's context information (e.g., c-RSU's link-local address and prefix called prefix1), and the vehicle's context information (e.g., the vehicle's global IP address and MAC address). After n-RSU receives the PBA message including the handoff context from MA2, it sets up a bi-directional tunnel with MA2, and generates RA messages with c-RSU's context information. That is, n-RSU pretends to be a router belonging to Subnet1. When the vehicle receives RA from n-RSU, it can maintain its connection with its corresponding node (i.e., CN1). Note that n-RSU also sends RA messages with its domain prefix called prefix2. The vehicle configures another global IP address with prefix2, and can use it for communication with neighboring vehicles under the coverage of n-RSU.¶
If c-RSU and n-RSU are adjacent, that is, vehicles are moving in specified routes with fixed RSU allocation, the procedure can be simplified by constructing the bidirectional tunnel directly between them (cancel the intervention of MAs) to alleviate the traffic flow in MA as well as reduce handoff delay.¶
Figure 6 shows the vehicle handoff within two prefix domains (as two multi-link subnets) with DMM [RFC8885]. If c-RSU detects that the vehicle is going to leave its coverage and to enter the area of an adjacent RSU (n-RSU) belonging to a different prefix domain, it sends a PBU message to inform n-RSU that the vehicle will enter the coverage of n-RSU along with handoff context such as c-RSU's context information (e.g., c-RSU's link-local address and prefix called prefix1), and the vehicle's context information (e.g., the vehicle's global IP address and MAC address). After n-RSU receives the PBA message including the handoff context from c-RSU, it sets up a bi-directional tunnel with c-RSU, and generates RA messages with c-RSU's context information. That is, n-RSU pretends to be a router belonging to Subnet1. When the vehicle receives RA from n-RSU, it can maintain its connection with its corresponding node (i.e., CN1). Note that n-RSU also sends RA messages with its domain prefix called prefix2. The vehicle configures another global IP address with prefix2, and can use it for communication with neighboring vehicles under the coverage of n-RSU.¶
This document shares all the security issues of Vehicular ND [ID-IPWAVE-VND], Proxy MIPv6 [RFC5213], and DMM [RFC7333][RFC7429][RFC8885].¶
This work was supported in part by Institute of Information & Communications Technology Planning & Evaluation (IITP) grant funded by the Korea Ministry of Science and ICT (MSIT) (2020-0-00395-003, Standard Development of Blockchain based Network Management Automation Technology).¶
This work was supported in part by the MSIT under the ITRC (Information Technology Research Center) support program (IITP-2022-2017-0-01633) supervised by the IITP.¶
This document is made by the group effort of IPWAVE working group. Many people actively contributed to this document, such as Carlos J. Bernardos and Russ Housley. The authors sincerely appreciate their contributions.¶
The following are coauthors of this document:¶
Zhong Xiang - Department of Electrical and Computer Engineering, Sungkyunkwan University, 2066 Seobu-ro Jangan-gu, Suwon, Gyeonggi-do 16419, Republic of Korea. EMail: xz618@skku.edu¶
Hyeonah Jung - Department of Computer Science and Engineering, Sungkyunkwan University, 2066 Seobu-ro Jangan-gu, Suwon, Gyeonggi-do 16419, Republic of Korea. EMail: hyeonah214@skku.edu¶
The following changes are made from draft-jeong-ipwave-vehicular-mobility-management-09¶