The IEEE 802.16m Reference Model

Figure below illustrates the IEEE 802.16 reference model. The data link layer of IEEE 802.16 standard comprises three sub-layers. The service-specific convergence sub-layer (CS) provides any transformation or mapping of network-layer data packets into MAC SDUs. On the transmitter side, the CS receives the data packets through the CS Service Access Point (SAP) and delivers MAC SDUs to the MAC Common Part Sub-layer (MAC CPS) through the MAC SAP. This includes classifying network-layer SDUs and associating them with the proper MAC Service Flow Identifiers (SFID) and Connection Identifiers (CID). The convergence sub-layer also includes payload header suppression function to compress the higher-layer protocol headers. Multiple CS specifications are provided for interfacing with various network-layer protocols such as Asynchronous Transfer Mode (ATM)i and packet-switched protocols such as IP or Ethernet. The internal format of the CS payload is unique to the CS, and the MAC CPS is not required to understand the format of or parse any information from the CS payload.  

The IEEE 802.16 reference model

The MAC CPS provides the core MAC functionality of system access, bandwidth allocation, connection establishment, and connection maintenance. It can receive data from the various convergence sub-layers, through the MAC SAP classified into particular MAC connections. An example of MAC CPS service definition is given in reference. The Quality of Service (QoS) is further applied to the transmission and scheduling of data over the physical layer. 

The MAC also contains a separate security sub-layer providing authentication, secure key exchange, and encryption. The user data, physical layer control, and statistics are transferred between the MAC CPS and the Physical Layer (PHY) via the PHY SAP which is implementation-specific. The IEEE 802.16 physical layer protocols include multiple specifications, defined through several amendments and revisions, each appropriate for a particular frequency range and application.

The IEEE 802.16 compliant devices include mobile stations or base stations. Given that the IEEE 802.16 devices may be part of a larger network, and therefore would require interfacing with entities for management and control purposes, a Network Control and Management System (NCMS) abstraction has been introduced in the IEEE 802.16 standard as a “black box” containing these entities. The NCMS abstraction allows the physical and MAC layers specified in the IEEE 802.16 standard to be independent of the network architecture, the transport network, and the protocols used in the backhaul, and therefore would allow greater flexibility. The NCMS entity logically exists at both BS and MS sides of the radio interface. Any necessary inter-BS coordination is coordinated through the NCMS entity at the BS. An IEEE 802.16 entity is defined as a logical entity in an MS or BS that comprises the physical and MAC layers on the data, control, and management planes.

The IEEE 802.16f amendment (currently part of IEEE 802.16-2009 standard) provided enhancements to IEEE 802.16-2004 standard, defining a management information base (MIB), for the physical and medium access control layers and the associated management procedures. The management information base originates from the Open Systems Interconnection Network Management Model and is a type of hierarchical database used to manage the devices in a communication network. It comprises a collection of objects in a virtual database used to manage entities such as routers and switches in a network.

The IEEE 802.16 standard describes the use of a Simple Network Management Protocol (SNMP),ii i.e., an IETF protocol suite, as the network management reference model. The standard consists of a Network Management System (NMS), managed nodes, and a service flow database. The BS and MS managed nodes collect and store the managed objects in the form of WirelessMAN Interface MIB and Device MIB that are made available to network management system via management protocols, such as SNMP. A Network Control System contains the service flow and the associated Quality of Service information that have to be provided to BS when an MS enters into the network. The Control SAP (C-SAP) and Management SAP (M-SAP) interface the control and management plane functions with the upper layers. The NCMS entity presents within each MS. The NCMS is a layer-independent entity that may be viewed as a management entity or control entity. Generic system management entities can perform functions through NCMS and standard management protocols can be implemented in the NCMS. If the secondary management connection does not exist, the SNMP messages, or other management protocol messages, may go through another interface in the customer premise or on a transport connection over the air interface. Figure 3-4 describes a simplified network reference model. Multiple mobile stations may be attached to a BS. The MS communicates to the BS over the air interface using a primary management connection, basic connection or a secondary management connection. The latter connection types have been replaced with new connection types in IEEE 802.16m standard

CSN-anchored Mobility

The CSN-anchored mobility refers to mobility across different ASNs alternatively to mobility across different IP subnets, and thereby requires network layer mobility management. The mobile IP protocols are used to manage mobility across IP subnets, and to enable CSN-anchored mobility. This section describes mobile IP based macro-mobility between the ASN and CSN across R3 reference point. In the case of IPv4, this implies re-anchoring of the current FA to a new FA, and the consequent binding updates (or MIP re-registration) to update the upstream and downstream data forwarding paths. In CSN-anchored mobility, the anchor mobile IP FA of the MS is changed. The new FA and CSN exchange messages to establish a data forwarding path. The CSN-anchored mobility management is established between ASN and CSN that are in the same or different administrative domains. The mobility management may further extend to handovers across ASNs in the same administrative domain. The procedures for CSN-anchored mobility management and the change of MS point of attachment to the ASN may not be synchronized. In this case, the procedures may be delayed relative to the completion of link layer handover by the MS. 

In an intra-NAP R3 mobility scenario, an MS is moving between FAs within a single NAP domain. The R3 mobility event results in a handover between two FAs, thereby relocating the ASN R3 reference anchor point in the NAP. Note that R3 mobility does not automatically terminate or otherwise interfere with idle/sleep operation of the MS. The CSN-anchored mobility accommodates the scenario in which the MS remains in idle state or sleep mode until it is ready to transmit uplink traffic or is notified of downlink traffic by the serving BS. In all non-roaming scenarios, the HA is located in the CSN of H-NSP. For roaming scenarios, the HA is located in the CSN of either the H-NSP or V-NSP, depending on roaming agreement between H-NSP and V-NSP, user subscription profile and policy in H-NSP. The CSN-anchored mobility within a single NAP administrative domain does not introduce significant latency and packet loss. A make-before-break handover operation (i.e., when a data path is established between the MS and target BS before the data path with the serving BS is broken) is feasible within the same NAP administrative domain. To accomplish this procedure, the previous anchor FA maintains data flow continuity while signaling to establish the data path to a new anchor FA. The PMIP procedures do not require additional signaling over-the-air or additional data headers to perform CSN-anchored mobility. The CSN-anchored mobility activities are transparent to the MS. The MS uses Dynamic Host Configuration Protocol (DHCP) for IP address assignment and host configuration. DHCP is a network application protocol used by devices to obtain configuration information for operation in an IP network. This protocol reduces system administration workload, allowing devices to be added to the network with minimal user intervention 

WIMAX Mobility Management

The WiMAX network architecture supports two types of mobility: ASN-anchored mobility (intra-ASN) and CSN-anchored mobility. ASN-anchored mobility refers to a scenario where a mobile terminal moves between two base stations belonging to the same ASN while maintaining the same foreign agent at the ASN. The handover in this case utilizes R6 and R8 reference points. The CSN-anchored mobility refers to an inter-ASN mobility scenario where the mobile station moves to a new anchor foreign agent and the new FA and CSN exchange signaling messages to establish data forwarding paths. The handover in this case is performed via R3 reference point with tunneling over R4 to transfer undelivered packets. 

Figure below illustrates three different mobility scenarios supported in WiMAX networks. When the mobile station moves from positions 1 to 2, or 1 to 3, an ASN-anchored mobility through R8 or R6 reference points, respectively, is implied, whereas moving from position 1 to 4 involves a CSN-anchored mobility scheme though R3 reference point.