Wimax standards and Security The Wimax

TABLE 11.1
Generic MAC Header Fields

Name	Length (Bits)	Description
CI	1	CRC indicator
CID	16	1 = CRC is included in the PDU by appending it to the payload after encryption if any 0 = No CRC is included Connection identifier
EC	1	Encryption control
		0 = Payload is not encrypted 1 = Payload is encrypted
EKS	2	Encryption key sequence
		The index of the traffic encryption key (TEK) and initialization vector used to encrypt the payload. This field is only meaningful if
		the EC field is set to 1
HCS	8	Header check sequence
		An 8-bit field used to detect errors in the header
HT	1	Header type
		Shall be set to zero
LEN	11	Length
		The length in bytes of the MAC PDU including the MAC header
Type	6	and the CRC if present This field indicates the subheaders and special payload types present in the message payload

The bandwidth request PDU, shown in Figure 11.4, has no payload and consists of only the header. It is 6 bytes in length and consists of 8 fields, which are defined in Table 11.3. Like the GMH, the bandwidth request header is encoded from the HT field on.

TABLE 11.2
Type Encodings
Type Bit Value

#5 (MSB) Mesh subheader 1 = present

0 = absent
#4 ARQ feedback payload 1 = present
0 = absent
#3 Extended type
Inidicates whether the present packing or fragmentation Subheaders are extended
1 = extended
0 = not extended
#2 Fragmentation subheader
1 = present
0 = absent
#1 Packing subheader
1 = present
0 = absent
#0 (LSB) Downlink: FAST-FEEDBACK allocation subheader Uplink: grant management subheader
1 = present

MSB
0 = absent

HT = 1(1)	EC = 0 (1)	Type (3)	BR MSB (11)
BR LSB (8)				CID MSB (8)
CID LSB (8)				HCS (8)

LSB
FIGURE 11.4
Bandwidth request header.

2. 
   PMP
   

In PMP mode, multiple subscriber stations connect to a single base station. Each subscriber station is uniquely defined by a 48-bit universal MAC address. It is used during the initial ranging process and during the authen- tication process so the base station and subscriber station can verify each other’s identity [5].


TABLE 11.3
Bandwidth Request Header Fields
Name Length (Bits) Description

BR 19 Bandwidth request

The number of bytes of uplink bandwidth requested by the subscriber station. The bandwidth request is for the CID. The request shall not include any PHY overhead

CID	16	Connection identifier
EC	1	Always set to zero
HCS	8	Header check sequence
		An 8-bit field used to detect errors in the header
HT	1	Header type = 1
Type	3	Indicates the type of bandwidth request header

When a subscriber station first connects to a base station, two pairs of man- agement connections are created between the subscriber station and the base station. An optional third pair of management connections may be created. Each pair consists of one uplink and one downlink connection, identified by a 16-bit connection ID (CID). Short, time-urgent management messages are sent over the basic connection. Longer, delay-tolerant management messages are sent over the primary management connection. Standards-based (i.e., DHCP, TFTP, SNMP) messages are sent using the secondary management connection.
Base stations do not have to coordinate their transmissions with other stations. They simply divide time into uplink and downlink transmission periods using TDD. Downlink messages are generally broadcast. A downlink map (DL-MAP) message can be used to define access to the downlink infor- mation by defining burst start times to subscriber stations. If a DL-MAP does not explicitly indicate a portion of the downlink for a specific subscriber sta- tion, then all subscriber stations capable of listening will listen. The subscriber stations will check the CIDs of the PDU and keep only the ones addressed to them.
Uplink transmissions to the base station are shared among subscriber sta- tions and are on a demand basis. Subscriber stations use an uplink map (UL-MAP), which is obtained from the base station, to determine when it can transmit. Four different types of uplink scheduling mechanisms are used to control contention between users and tailor the delay and bandwidth require- ments of each user application. These are implemented using unsolicited bandwidth grants, polling, and contention procedures. Performance can be optimized by using different combinations of these bandwidth allocation techniques.

3. 
   Mesh
   

In mesh mode, subscriber stations can transmit to each other directly, allowing traffic to be routed through subscriber stations if two nodes can- not communicate directly. The advantage of mesh mode is it can provide NLOS communication for stations using higher frequency bands. This is accomplished by marking a node as a mesh base station if it has a direct con- nection to backhaul services outside the mesh network. Otherwise it is marked as a mesh subscriber station. Traffic can then flow from mesh subscriber stations to mesh base stations, then out of the mesh network and vice versa [5]. Similar to PMP mode, each node is uniquely defined by a 48-bit universal MAC address. It is used during the network entry process and the authentica- tion process where the entry node and the network verify each other’s identity. Once a node is authorized to the network, it requests a 16-bit node identifier (node ID) from the mesh base station. This node ID is used to identify nodes
during operation.
Nodes view other stations in its mesh network in three different ways. Neighbors are stations to which the node has a direct link, which are consid- ered to be “one hop’’ away. A neighborhood consists of all the neighbors of a node. Finally, an extended neighborhood contains all the neighbors of the neighborhood in addition to the neighborhood itself.
All communications within a mesh network are in the context of a link. 8-bit link identifiers (link IDs) are used to address nodes in the local neighborhood. Each link established between a node and its neighbor shall be assigned a link ID. As neighboring nodes establish new links, link IDs are communicated during the link establishment process. All data transmissions between the two nodes use the same link.
Mesh mode uses two types of scheduling, distributed and centralized. In distributed scheduling, all the nodes must coordinate their transmissions in their extended neighborhood. This can be accomplished by having every node broadcasting its schedule (available resources, requests, and grants) to all its neighbors. Schedules may also be established by directed uncoordinated requests and grants between two nodes. Before transmitting, a node must ensure that it will not cause collisions with the transmissions scheduled by any other node in its extended neighborhood.
In centralized scheduling, resource request from all the mesh subscriber stations within a certain hop range is gathered by the mesh base station. The base station determines the amount of resources it wishes to grant on each link in the network, and communicates the grants to all the mesh subscriber stations in the hop range.

4. 
   QoS
   

WiMAX was designed with QoS in mind, to provide low latency for delay- sensitive services and data prioritization. QoS support resides within the MAC layers of the base station and subscriber stations. The base station contains a packet queue for each downlink connection. It uses the QoS

parameters and the status of the queues to determine which queue to use for the next SDUs to be sent. The subscriber station has similar queues for uplink connections [6].

Bandwidth is granted to the subscriber stations from the base stations when it is needed. Subscriber stations can request bandwidth in a few different ways. Using unsolicited granting, during the setup of an uplink connection, subscriber stations request a fixed amount of bandwidth on a periodic basis. Once the connection is complete the subscriber stations cannot request any more bandwidth. The base station can use broadcast polls to determine if sub- scriber stations need bandwidth. An issue arises when two or more stations respond to the same poll causing a collision. After collision, nodes follow an exponential backoff algorithm and wait to respond again. Bandwidth requests can also be piggybacked on a PDU sent from the subscriber station.
The base station’s uplink scheduler uses the bandwidth requests to esti- mate the remaining backlog at each uplink connection. It uses this knowledge and the set of QoS parameters to determine future uplink grants. While the bandwidth requests are per connection, the base station grants uplink capac- ity to each subscriber station as a whole. Therefore, the subscriber station also implements a scheduler within its MAC to allocate its uplink bandwidth between its connections.

3.