The proxy Mobile Internet Protocol (PMIP) is an IP mobility management network protocol introduced by IETF in order to assist in the support of IP mobility in a low latency, high data-rate across a heterogeneous network with various access technologies. It gives mobile terminal access to mobility without involvement in the management of their IP mobility signalling (Sanchez, M. I. et al; 2013) and hence depends on an additional software and hardware implemented on the mobile terminal (Ruckus wireless; 2013).
Mobility management is supported in PMIP by two key elements: the Mobile Access Gateway (MAG) and the Local Mobility Anchor (LMA). The LMA in the mobile core is responsible for maintaining the location of the mobile terminal in the Localized Mobility Domain (LMD) and forwarding the data traffic of the mobile terminal by maintaining an IPv6-in-IPv6 channel with the MAG which is in the core network.
In PMIP, the mobile terminal always maintains its IP address as it moves from one point to within the LMD managed by an LMA. Similarly, “the operation of PMIPv6 does not require the MN to implement any modification or extra software in its layer-3 stack, although it may require the assistance of some layer-2 mechanisms to work more efficiently. These mechanisms are known as link-layer triggers, and are required to quickly detect a change of layer-2 Point of Attachment (PoA)” (Sanchez, M. I.; 2013).
PMIP is one of the technologies that perform localized mobility management which allows a mobile terminal (MT) to move from one access router to another within the same network in a transparent way. It allows for the reduction in mobility signalling traffic and the improvement of handover performance (Enable book, 2008).
Mobility in GSM Networks
Mobility in GSM can realized through the switching capability and data subscription separation. That is, by grouping some PLMNs which provide communication capabilities that the network provider can provide, and passing information among them. The PLMN consist of MSC and GMSC and are not specific to a certain group. The MSC which contain the subscribers’ information has a switching capability that may be used by the subscriber using the PLMN and for subscribers’ mobility within a network. As far as the mobile terminal is connected to the Base Station (BSS), the information of the subscriber is contained in the MSC/VLR. This information is kept in the MSC even when the mobile terminal is disconnected from the BSS for a certain period of time and when it moves from one MSC to the other, the HLR will instruct the previous MSC to purge the subscription information of the subscriber and send the information to the present MSC (Eberspacher, J.; Bettstetter, C.; and Vogel, H-J.; 2009).
Mobility in 3G Networks (UMTS)
The UMTS signal protocol is divided into two strata of access strata (AS) and non-access strata (NAS). The NAS, which is located at the upper layer, is responsible for: Connection Management (CM) for handling circuit-switched calls control, short message services and supplementary services; Session Management (SM) responsible for packet-switched calls; Mobility Management (MM) which handles mobility functions such as location update in the circuit-switched domain; and GPRS Mobility Management (GMM)which is responsible for handling mobility functions such as route update in the packet-switched domain.
The Access Stratum is made up of the physical layer (PHY) in layer 1; Radio Link Control (RLC) and Medium Access Control (MAC) in the second layer; and Radio Resource Control (RRC).
Mobility management in UMTS is mostly done in three major states. The three mobility management states of the mobile terminal (MT) exist. The mobility management states are the detached, connected and idle states.
At the detached state, the mobile terminal is unreachable by the network because it is not attached to the UMTS. In the idle state, the mobile terminal is attached to the UMTS network and mobility management have been established by both the mobile terminal and the SGSN by performing detach/location update and paging procedures respectively. Here, the mobile terminal is tracked by the SGSN at the routing area (RA) level. In the connected state, packet-switched signalling connection is established between the mobile terminal and the SGSN. Unlike in the idle state, the mobile terminal is tracked by the SGSN with precision of the routing area and the serving RNC relocation which is responsible for tracking call-level is executed here.
Mobility in WiMAX
Mobility Management in the WiMAX architecture are designed to achieve minimal handover latency and packet loss. The WiMAX foundation basing on the procedure specified by the IEEE 802.16 standard, has designed network messaging and procedures to provide the network support of mobility.
WiMAX network architecture uses Mobile IP Protocols and special protocols defined by WiMAX NWG to handle mobility. The MIP makes it possible to use off the shelf components such as Home Agents (HA) and the WiMAX specific protocols are used to provide optimization and flexibility in handling mobility. For a given mobile terminal, the MIP HA which resides in the CSN and one or more Foreign Agents resides in each ASN. The mobile terminal’s information is transported through the MIP tunnel which ends at the FA in the ASN. On termination of the MIP tunnel at the FA, the data path function (DPF) which is the WiMAX specific protocol takes over and transport the information from the FA to the operating base station (BS) to which the mobile station is attached.
Hence, mobility management in the WiMAX network can be classified as a two-tiered solution: ASN-Anchored and CSN-Anchored. ASN-anchored relates to handover event without the MIP tunnel termination point while the CSN-anchored mobility management are performed on all handover activities and include several localized optimizations such as extending the data path from a previous serving ASN entity to the new ASN, thereby avoiding the effect of Layer-3 handover delays when applicable (Shantidev, M.; Venkatachalam, M.; and Yang, X.; 2008).
Mobility in TETRA
An optimal mobility management in TETRA will provide a dynamic and critical public safety and disaster relief network will provide a consistent, seamless and secure communication services in cases of emergencies.
Annalisa, D.; and Marco, P. (2008), suggest a vertical handover (VHO) solution able to optimize mobility management over a heterogeneous network which is specifically designed to provide public safety and disaster relief organisation using TETRA and TETRA 2 systems with the capability to exploit Wi-Fi and WiMAX broadband access technologies in order to enjoy advanced services.
The concept supports both low-level and high-level mobility by pursuing a cross-layer approach that lets MIP and SIP with SBC collaborate in a complimentary rather than competing way, thus avoiding redundancy. It balances the traffic load among different networks and handles real time traffic with particular respect.
Mobility in Mobile Ad hoc Network
Mobility management in mobile ad hoc network (MANET) takes place in the Local Mobility Anchor (LMA) which will register and update the location of the Mobile terminal inside the Localized Mobility Domain (LMD). The Access Routers (AR) alerts the LMA when a mobile terminal moves from one router to the other in order to update its location. However the access router cannot identify a mobile terminal when they are more than one hop apart because MANET is a multi-hop network (Sergento, S.; et al; 2008).
MANET needs an adaptation in order to work in a LMD which relates to the ability of the access router to detect the mobility of mobile terminals. On arrival in a network, the mobile terminal alerts the AR (or gateway) of its location which in turn activates the mobility protocol. Once it is identified by the mobility protocol via bootstrapping, the mobility protocol can operate in the same way than in infrastructure network.
Mobility Management in Heterogeneous Networks
As mentioned earlier, in a heterogeneous wireless network there is a need for a mobility management at layers above the data-link layer in “order to take advantage of all available technologies at a certain moment and a certain place”(Anderson, K.; 2012: 32).
Munoz, P.; Barco, R.; Laselva, D.; and Mogensen, P.; (2013); Introduces different mechanisms for steering traffic by clarifying the techniques that may adjust mobility parameters and the challenges arising from particular deployment of heterogeneous network including LTE deployment; And proposed a fuzzy-based algorithm that enhances network constraints for traffic steering.
In the idle mode, the AP-based cell reselection algorithm indicates that the Aps of the HetNets allows the service providers to control the distribution of users across the network layers. In the connected mode, the inter-RAT HO algorithm on the other hand shows that the adjustment of the associated parameters may depend on the small cell location within the macro cell. Lastly, the SON algorithm which is based on reinforcement learning and fuzzy logic shows that “it can automatically adapt to context variations in order to significantly improve user satisfaction in Heterogeneous networks.
Thakur, G.S.; and Helmy, A.; (2013); investigates the capability of existing mobility models by taking into consideration the various aspects of mobility behavior, as well as network protocol performance and then proposed the collective behavior based on realistic aspects of human mobility (COBRA) which attempts to explicitly takes into consideration individual, pair-wise and group mobility.
The model COBRA is capable of spanning the mobility metric space and matches realistic traces with protocol performance, thus reducing “the gap between current models and human behavioral mobility modeling” (Thakur, G.S.; and Helmy, A.; 2013).
Sanchez, M.I. et al; (2013); proposes a network-based integrated mobility framework for wireless optical broadband access network (WOBAN) which is based on the proxy mobile IPv6 (PMIPv6) and IEEE 802.21 (MIH) mobility management protocols. PMIPv6-WOBAN aimed at providing an enhanced the mobility of “users with respect to the overall network resources, both at the wireless access and optical distribution parts; remove the overhead of IP-in-IP tunneling between the PMIPv6 entities, and perform an efficient bicastng during handover process to minimize packet loss”(Sanchez, M.I. et al; 2013).
The above objectives were achieved by mapping the PMIPv6 framework and IEEE 802.21 MIH services into the hierarchical structure of the WOBAN’s Passive Optical Network (PON) by collocating the Local Mobility Anchor (LMA) with the Optical Line Terminal (OLT), and the Mobile Access Gateways (MAGs) with the Optical Network Units (ONUs) which controls a set of heterogeneous wireless Access Points. The OLT-LMA node is able to combine information on mobile mobility and traffic statistics obtained from the ONU-MAGs, thereby initiating handovers of MTs due to ONU and AP overload; hence making the integrated framework provides an enhanced use of resources. Also, the framework contains a number of optimizations that prevent packet loss and providing seamless handover by the single-hop, point-to-multipoint topology of an Ethernet PON (EPON) which avoids the overhead of maintaining tunnels between LMA and its MAGs; and enables the use of multicast EPON bicasting during mobility management. Finally, the PMIPv6-WOBAN framework provides mobility between different networks on the Localized Mobility Domain (LMD).
Distributed Mobility Management (DMM)
Most of the currently standardized IP mobility management protocols such as Proxy Mobile IPv6 (RFC5213) rely to a certain extent on a centralized mobility anchor entity which is in charge of the mobility control and the forwarding of users’ data. They offer mobility support at the cost of handling operations at a cardinal point, the mobility anchor, and burdening it with data forwarding and control mechanisms for a great amount of users. As stated in [RFC7333], centralized mobility solutions are prone to several problems and limitations: longer (suboptimal) routing paths, scalability problems, signalling overhead (and most likely a longer associated handover latency), more complex network deployment, higher vulnerability due to the existence of a potential single point of failure, and lack of granularity on the mobility management service (i.e., mobility is offered on a per-node basis, not being possible to define finer granularity policies, as for example per-application). This in turn makes the current mobility protocols prone to several problems and limitations (H. Chan, 2011) [RFC7333] such as single point of failure, routing in a non-optimal route (suboptimal), overloading of the centralized data anchor point due to the data traffic increase, low scalability of the centralized route and context management, scalability problems, signalling overhead (and most likely a longer associated handover latency), more complex network deployment, higher vulnerability due to the existence of a potential single point of failure, and lack of granularity on the mobility management service.
This has prompted researchers and mobile operators to look for different mobility management methods which are more distributed in nature, and that allow to enable mobility on demand for particular types of traffic. This effort brought about what is known as Distributed Mobility Management (DMM) (Fabio Giust, et al; 2009).
One of the major requirements for distributed mobility management (DMM) as defined by (RFC7333) is to enable traffic to avoid traversing single mobility anchor far from the optimal path by addressing the scalability issues derived from a centralized mobility management (CMM) deployment and providing mobile nodes with local anchors.
DMM whose concept is based on the distribution of mobility anchors towards the access networks to provide mobile nodes with local anchors and enable optimized routing of traffic above anchor level to any kind of serving point, according to Luca Valtulina (2013) offers to operators a more efficient network deployment driven by a distributed placement of core network entities close to the edge (access) of the network.
Distributed mobility management aims at solving the centralized mobility anchor problems of the traditional mobility management protocol. The benefit of DMM solution is that the data plane traffic does not need to traverse the centralized anchoring point.
Current IP mobility solutions, standardized with the names of Mobile IPv6 [RFC6275], or Proxy Mobile IPv6 [RFC5213], just to cite the two most relevant examples,
The purpose of Distributed Mobility Management is to overcome the limitations of the traditional centralized mobility management RFC7333] [RFC7429]; the main concept behind DMM solutions is indeed bringing the mobility anchor closer to the MN.
Distributed Mobility Management (DMM) allows network traffic to distribute among multiple mobility anchors which have mobility functions to solve the existing problems in current centralized mobility protocols. There are many DMM approaches extending network-based mobility protocols (e.g. Proxy Mobile IPv6).
In Proxy Mobile IPv6 (PMIPv6), they allow a mobile node to connect to PMIPv6 domain through different physical interfaces.
In DMM scenario, mobility anchors would be deployed in a distributed manner, and as specified in [RFC7333], one of the aims of DMM is to reduce the routing redundancy between mobile node and correspondent node, which means providing a more optimal communication path for application traffic between mobile node and correspondent node. To achieve routing optimization for specific application traffic, the basic idea is to make the traffic using IP address(s) anchored at current anchor, so that downlink traffic from correspondent node to mobile node will go directly to mobile node, but this routing optimization requirement brings a fact that mobile node has to change its IP address as it moving to a new anchor. Some application sessions can cope with the change of IP address either by application layer itself or by the function provided by other layers, e.g. transport layer; but for other application sessions, after IP address changed, the application session would be broken off totally.
So it’s reasonable to provide different network layer mobility support according to the need of application.