Yes.  WiMAX radios already support very robust QOS capabilities up to and incorporating asynchronous transfer mode (ATM) quality.  The MAC itself is configured to handle IP traffic, Ethernet and ATM natively.  The MAC was designed to even support future transport protocols not yet invented.  Links can be dynamically configured based on link conditions.  Basically, this dynamic configuration technique smoothens the balancing act between raw capacity and quality on the fly.  It should improve capacity or spectral efficiency a great deal.  The whole issue of QOS is becoming even more important as device capability (such as the iPhone) have revealed an apparent pent-up desire for consumers to use more rich media content such as video; a desire that was previously hidden by devices that made the experience too hard for consumers to utilize easily. 

There are a lot of elements of wireless transmission which affect the quality of signal---needs also vary depending on the type of data.  For example, VoIP can tolerate some errors, but must have low latencies (anything above 150 ms is problematic) to operate.  The packet sizes for VoIP are typically much smaller than for data.  When networks must handle blended traffic, the polling mechanism that chooses which radio can transmit with either a smaller VoIP packet or a larger data packet is crucial to ensure that data traffic is not optimized at the expense of voice.  Video transmission is similar.  Conversely, data packets do not need especially low latencies, but cannot endure transmission errors.

WiMAX partly accomplishes this by assigning variable length Protocol Data Units (PDU)s, which is basically the data packet size in the Physical Layer, that can be combined in bursts to reduce signaling overhead in the PHY layer.  This is called adaptive modulation and is a sharp contrast from the static modulation schemes of the past.  A similar technique is used for MAC signaling except they are called Service Data Units (SDU)s.  Several other techniques are used for reducing signaling transmissions and to improve the polling or communications between radios.  In the older 802.11b protocol for example, each radio and base station continues to signal and interact constantly with other radios---basically a carrier sense multiple access with collision detection (CSMA/CD) approach similar to Ethernet computer networks.  This unfortunately results in packet collision, packet loss and a great deal of inefficient cross talk in a static mode. 

WiMAX technology supports a variety of more efficient polling mechanisms that vendors and carriers can choose to use, including a defined contact cycle, grouping of radios into contact groups or even allowing customer radios to generate a brief signal indicating it needs a transmission cycle.  All of these aspects, which are intended to solve multiple problems, also result in improved QOS capabilities.  QOS is critical for delineating minimum bandwidth levels for VoIP sessions for example, as well as other leading edge IP services. 

Both common duplexing schemes are supported in WiMAX---those being FDD and TDD.  The frequency division duplexing (FDD) requires two parallel channels for send and receive.  This method is a well-understood holdover from cellular technology.  The newer time division duplexing (TDD) allows for dynamic and symmetric transmission of data across a single channel.  Where and when either should be used often depend on the frequency and the vendor's emphasis on particular strengths.  It is not unfair to suggest that TDD is more likely to be widely utilized by WiMAX product vendors.  Suffice it to say that multiple duplexing support adds significant flexibility to WiMAX---capabilities not before supported by broadband wireless technology.