Wimax standards and Security The Wimax

Background

1. 
   QoS Definition
   

QoS is the guarantee of the service-level performance for a data stream from a source to a destination [3]. Such an assurance, of course, shall not exceed the

Network device



Data stream 1 (low)



Data stream 2 (medium)



Data stream 3 (high)
Outgoing traffic

Incoming traffic


FIGURE 4.1
Need for QoS.
physical capacity of the transmission media. For example, we cannot guar- antee a throughput of 100 Mbps on a Cat-3 cable that supports only 10 Mbps. Also, we cannot guarantee one-way delay faster than the speed of light on any long-haul network. The need for QoS arises when there are multiple data streams competing for the limited physical capacity of the transmission media or network devices (see Figure 4.1). In the case of WiMAX, the limiting resource is the radio frequency bandwidth. When there are multiple data streams competing to use the same frequency bandwidth, a QoS policy is needed to determine which data stream has the priority to use the air interface. This QoS policy depends on the user applications that are characterized by QoS performance metrics. For example, an e-mail application does not need any guarantee except for reliable delivery of the data. A VoIP application needs guarantee of low latency. A video-streaming application can afford a long delay but requires relatively high bandwidth. The following elements are required to implement QoS on a network:

1. 
   QoS performance metrics: QoS is a mechanism to assure network performance as defined by a set of metrics associated with each data stream. Examples of performance metrics are delay, throughput, jitter, and packet loss.
   
2. 
   Request and grant: This is also known as admission control. In the case of WiMAX, the BS is the central control point. An SS requests a connection with certain QoS parameters. If the network does not have the resource, the request will be rejected. If the network has sufficient resource, the BS will check if the SS is authorized to use the resource. After authorization, the BS will guarantee the service throughout the connection.
   
3. 
   Traffic shaping: For an incoming packet, the network device needs to determine how to classify the packet and whether to send the packet. If the packet delivery is not guaranteed and the network is congested, the packet could be dropped. Otherwise, the packet enters into a priority queue and waits for the scheduler to deter- mine its delivery. The IEEE 802.16 standard does not require traffic shaping as the air interface would not drop packets. The traffic
   

shaping on the wireline side of the device is outside the scope of the standard.

4. 
   Scheduling policy: A QoS-enabled device has multiple priority queues for different classes of services. The scheduling policy is to determine how and when to process packets in the priority queues. A scheduling policy could use a round-robin method to process packets in each priority queue and allocate more resources for high- priority queues. Another scheduling policy could be to process packets in a low-priority queue only when all high-priority queues are empty.
   

2.