Previous Work

Previous work that is closely related to this project involves experience with similar software simulation tool to the one that is used for the completion of the practical part of the project. OPNET simulator was used for the set up of a small network and its division into subnets. The project was completed as a lab assignment in the City University laboratories. It was divided into two main lab sessions each of them explained as follows.

The aim of first lab session was to demonstrate the need for implementation of switches in the design of the local area networks. Two 16-station LANs were designed, using a hub for the first one; two hubs and a switch for the second one. The main task of the second laboratory was to demonstrate the basics of designing a network, taking into consideration the users, services & location of hosts.

Using OPNET software, the concepts of networking were easily implemented and tested against the criteria set for satisfactory performance of the network model. The visual representation of the obtained results was used for a detailed analysis of the complexity of the network and its services provided to the users.

1 Technical Overview of the WiMAX Standard

1.1 Spectrum of the Standard

The IEEE 802.16 standard is initially designed to provide a flexible, cost-effective, standards-based last-mile broadband connectivity to fill in the broadband coverage gaps that are not currently served by the ‘wired' solutions such as DSL. The advanced versions of the standard are aiming to create new forms of broadband services both with high speed and mobility. The IEEE 802.16 standard envisages the use of a wide range of frequencies from 2 to 66 GHz. However, the WiMAX Forum has focused on the use of 2 to 10GHz as the operating frequencies.

A graphical representation of the WiMAX spectrum bands is listed below. It includes a comparison with other wireless standards; as well as indication of the regions using the listed bands.

WiMAX is a technology that supports the delivery of last-mile wireless broadband access as an alternative to cable and DSL. WiMAX provides fixed, nomadic, portable and ultimately mobile wireless broadband connectivity without the need for line-of-sight with a base station. The design of a WiMAX network is based on the following major principles:

Spectrum: to be deployed in both licensed and unlicensed spectra.

Topology: supports different RAN topologies.

Interworking: autonomous RAN architecture that enables faultless incorporation and interworking with Wi-Fi, 3GPP and 3GPP2 networks and existing IP core networks (e.g. DSL, cable, 3G) using IP-based interfaces.

IP connectivity: supports a mix of IPv4 and IPv6 network interconnects.

Mobility management: opportunity to expand the fixed access to mobility and broadband multimedia services delivery.

WiMAX has defined two MAC classification profiles -the ATM and the IP. They have also defined two primary PHY system profiles:

25 MHz-wide channel (in the US) in the 10-66 GHz range.

28 MHz-wide channel (in Europe) in the 10-66 GHz range.

IEEE 802.16 standard is designed to develop as a set of air interfaces standards for WMAN based on a universal MAC protocol but using the physical layer specifications, which one dependent on the range of use and the related regulations. The IEEE 802.16 working group designed a flexible MAC layer and associated physical layer for 10-66 GHz.

It is more important to recognize certain factors that make some frequencies more suitable for use in WiMAX systems, both fixed and mobile [1new].

1.1.1 Path Loss

The first important factor is the operating frequency and path loss, which may arise from three basic factors:

Free space path loss (FSL)

It is defined by using the following formula (Equation 1):

FSL=10log(4πDFC)2 (1)

D=distance from transmitter; F=frequency; C=speed of light.

The path losses increase with the square of the frequency. For example, the path loss at 2 GHz frequency band is about 12 dB higher than the band at 0.5 GHz. This effect may be compensated with the antenna size.

The increase in FSL with the square of the frequency requires the cell sizes at higher frequencies to be smaller in order to maintain link margins. High frequencies such as 5.8 GHz and above are best suited for line of sight environment. In NLOS conditions, a link at 5.8 GHz would support NLOS customer premises equipment at distances less than a kilometre.

Loss due to NLOS operations

In urban environments, WiMAX systems operate in a NLOS manner, there is a loss in received signals which depends on the reflected signal strengths. Frequencies above 10 GHz are treated separately. In general, lower frequencies such as 800-2000 MHz have better performance for NLOS than the higher bands. Due to the reception of waves reflected from many objects, the signal strength in most NLOS conditions varies sharply. Ground propagation models are required for path loss analysis and as a result additional margin needs to be given for the loss expected. Hence, multiple antenna techniques with spatial diversity are used to improve the margins in NLOS conditions.

Loss caused by in-building penetration

This kind of loss depends largely on the type of wall and whether the indoor location has windows. They are not very frequency dependent, at least in the zone of consideration of 2 GHz to 4 GHz. In buildings, losses can vary from 2 dB - for a room with windows; to 6 dB - for a brick wall. An increase up to 10-12 dB can be expected if the indoor areas were built using metal materials.

1.1.2 Doppler Shift and Coherence Time

The second important factor is the Doppler Shift. It is an essential consideration for systems meant for mobile use. In the case of mobile WiMAX systems, which are meant to cater to vehicular speeds of up to 120 km/h, the effects of the Doppler shift are only relevant to the higher frequencies. It is given by the following formula (Equation 2):

Ds=(VFC)cosα (2)

Ds=Doppler shift; V=velocity of user; F=frequency; C=speed of light and α=angle between the incoming signal and direction of motion.

Frequency shifts need to be less than 10% of the subcarrier spacing in order to maintain correct timing between different mobile stations, which may be operating with a base station. Table 1 represents the relationships between Doppler shift with operating frequency and speed.

Frequency (MHz)

Doppler shift (Hz)

Coherence time (ms)

Symbol duration (ms)

























Table 1. Coherence time in ms is relevant to time synchronisation of mobile stations with the base station (times are in milliseconds)

WiMAX is a multicarrier transmission system based on OFDM. The uplinks in WiMAX (both FDD and TDD systems) operate in a TDMA mode. Each subscriber station is assigned its own time slot and the transmission must take place within the slot. Other devices have the right to transmit in the other slots as allotted in the frame of WiMAX. In other words, the coherence of timing between different devices is important, which must operate in synchronisation. The coherence time is defined as 1Ds and it is inversely proportional to the Doppler shift. An indicator of how the motion is affecting the connection between symbols from different devices, a comparison of coherence time and the symbol time in OFDM.

Hence, rather than considering a shift in the centre frequency, which is more likely for single-carrier systems, it is more suitable to consider the same shift in time in the subcarriers.

Reducing subcarrier spacing would have adversely affected the Doppler spread, whereas increasing the subcarrier spacing would have reduced the delay spread and consequently intersymbol interference and the data rate.

1.2 Architecture

In addition to the standard spectrum analysis and factors that are important for the appropriate selection of correct frequencies, the next part of the report presents the ideas of WiMAX network architecture and specifications.

1.2.1 Backhaul Solutions

In telecommunications, thebackhaulportion of the network comprises the intermediate links between thecore networks, orbackbone, of the network and the small sub networks at the "edge" of the entire hierarchical network. Based on principle explained above, one of the architecture models is based on fixed IEEE 802.16 equipment. Independent of the used version, the technology can be employed in a fixed infrastructure as shown in Fig. 2. In this setting, the deployment of point-to-point connections can span tens of kilometres.

WiMAX backhauling is defined by the infrastructure between the WiMAX ASN-GW and the base stations. [2 new] The control of the wireless network towards the client devices, as well as the transport traffic to the operator CSN is aggregated by the base stations. Fig. 3 is another example of how the system may be structured. A critical achievement factor in the deployment of the WiMAX systems is the numerous options the service provider can choose from.

As the variety of options for the service providers were mentioned, there is time to focus on a topic much more related to the economic point of a network system. In particular, how the network will be develop in terms of cost and charges and they are agreed upon. A list of the important factors is presented as follows:

Fixed costof providing a network infrastructure.

Non-fixed cost of connection to the network- typically paid by the user in the form of connection cost.

Cost of increasingthe network's capacity. Users who want to reschedule their transmission during peak times should not be charged for the growth of the network's capacity.

Incremental cost of sending an extra packet. This cost should be very small or equal to zero without congestion, in view of the fact that the bandwidth of a broadband network is in general a shared resource.

Social costdefined as the extra delay which occurs to other users by the transmission of data.

Figure 4 shows the cost flow of a WiMAX network.

WiMAXpoint to multipoint



WiMAX MS/RS or WiMAX mesh BS


WiMAX MS/RS or WiMAX mesh BS + Wi-Fi router

WiMAXcell layout

Cell dimensioning

Number of WiMAX mesh BS in a cluster

Wi-FiCell layout

Number of AP in a single WiMAX cell Wi-Fi technology options

Dual interface boards (WiMAX/Wi-Fi)

Average subscribers per square mile

Data traffic assumptions for Wi-Fi users and WiMAX SS/RS (utility/demand function)

High QoS VoIP connections for Wi-Fi users and WiMAX SS

Average number of data connections per square mile / month

Backhaul capacity planning


WiMAX mesh to WiMAX main BS Wi-Fi routers to WiMAX BS

Radio capacity planning

Channel size

Modulation type

Frequency reuse

FCC limitations

Economic considerations

CAPEX: WiMAX BS, Wi-Fi routers, spectrum costs, site preparation, site installation, backbone network equipment

OPEX: Operation-administration-management costs, site Leases, equipment maintenance, customer acquisition

Network revenue - pricing model

Time unit fees for Internet subscribers and VoIP subscribers (flat rate pricing) On-demand Service (user-based pricing)

Charges: access, usage, congestion and QoS

Fig. 4 Cost flow diagram for WiMAX networks

1.2.2 Mobile WiMAX Network Solution

The second architecture model that is discussed is that of the mobile WiMAX. It is of great interest and use for the simulation project as it explains the principles behind the simulation that has been set.

A WiMAX network consists of an ASN in the lowest level and network itself is in fact, quite intuitive. [1 new] An ASN is comprised of a number of BS connected to an access network, where the ASN is connected to the external networks using ASN-GW. The functions of the ASN are represented in Fig. 5 and include a number of BS, access networks, and access gateways. The functions are discussed in details as follows:

  • Establishing of a connection with the MS, including PHY and MAC layer connectivity.
  • Providing handover and roaming services for the MS within the ASN
  • The user should be provided AAA facilities in conjunction with its home network; the ASN is capable of providing proxy AAA services
  • Relay facilities between the ASN and the external networks should also be present

Two functional entities are defined in ASN shown in Fig. 6:

The base stationconnects to the MS using the WiMAX PHY air interface. The functions of the BS are to maintain the air interface with the MS, providing the DHCP proxy and to maintain its status - idle or active. Fig. 6 is an illustration of a typical WiMAX BS. The uplink and downlink traffic scheduling as well as the QoS enforcement are provided by the BS, as the air interface is managed by the BS.

The access service network gateway is the second functional entity. The ASN-GW is the point where traffic from all BS is aggregated for interface to external networks. Usually, the ASN-GW is physically a router. Functions such as QoS management, AAA functionalities are also part of the aggregation point responsibilities. ASN-GW may also have control functions over the BS. The alternative architecture embeds those control functions in the BS. Fig. 7 represents the typical ASN functions.

If a comparison between WiMAX network architecture and a cellular network model such as GSM was to be done, similarities can easily be identified. The ASN-GW of the WiMAX network serves the functions that are provided by the SGSN; the access network is recognised as BSC in the GSM model, and the air interface is respectively the BST. Further reading and understanding of the GSM model architecture is recommended for the comparison between the two models. Generally, the WiMAX networks are defined to be much more open in building networks and they have greater level of functionalities in its network entities.

Three distinct entities comprise a WiMAX network and its application environment:

  • Network access provider - entity that operates one or more ASNs. Typically, it is a WiMAX operator that operates ASN in one or more areas.
  • Network service provider - provides connectivity and services to NAPs. NSP needs only to connect to other NSPs and expect all services to be delivered through these connections; NSPs provide connectivity to NAPs via CSN. They are also responsible for providing mobility between their own nodes as well as nodes from other NSPs.
  • Application service providers - provide services such as HTTP, video streaming, file download, e-mail, etc.

Fig. 8 introduces one more entity of the WiMAX network architecture - the connectivity service network. The diagram shows that CSNs have AAA servers which are policy functions for QoS. They provide connectivity to external networks such as managed IP networks or the public internet. Policy functions for each device, user and service on the network as well as the security and authentication frameworks are provided by the AAA servers. Authentication at multiple levels is a key feature of the WiMAX network architecture.

Additional key features of the mobile WiMAX network architecture include:

  • Both the core and the radio access networks should be based on IP; protocols based on IEEE802.16 and IETF
  • Open interfaces should be defined by fully-defined reference points
  • Support of fixed network, nomadic, or mobile usage with full migration path to mobility
  • A modular network architecture, which can grow based on usage requirements
  • Integration into different types of IP and non-IP networks (for example: ATM, TDM, and others)
  • Network core architecture is not based on support of any particular service, such as voice, data, or video; as a multi-service core network it supports QoS for each service and each connection
  • QoS is based on both policy functions and enforcement
  • The network architecture is relatively “flat”, which enables a WiMAX service to start with a single ASN; NWG has defined different ASN profiles for this purpose
  • Inter-networking is supported with 3GPP, 3GPP2, Wi-Fi, or wired networks using IETF protocols.

2 WiMAX Layers

After the spectrum and network analysis discussions, it is important to look deeper into WiMAX fundamentals. The following chapter has the aim to introduce the two main layers defined in the WiMAX standard, and to explain the related to them topics such as adaptive modulation, QoS, etc.

As the WiMAX belongs to the IEEE 802 group, then the bridging or layer-2 concepts should be mentioned. [2] The addressing is based on MAC addresses and the base station is perceived as a bridge. In order to implement all layer 2 functionalities, the device used should be a bridge-not a router. For the identification for nodes addressing is used; as long as the node is recognized by the network, the address is replaced by use of circuits with circuit identifier.

The WiMAX physical layer is based on orthogonal frequency division multiplexing. [2new] OFDM is a transmission scheme that enables high-speed data, video, and multimedia communications. It is used by numerous broadband systems - DSL, Wi-Fi, DVB-H, MediaFLO and WiMAX. OFDM is an efficient scheme used for high data rate transmission in a non-line-of-sight or multipath radio environment.

The MAC layer is based on connection-oriented principle and it is very similar to the ATM transport protocol, which connection uses a context that describes the mapping between the incoming flows and the underlying QoS. A station registers itself to the base station, negotiates the physical layer characteristics and then can communicate bidirectionally. A service flow defines the negotiated QoS for all matching packets (service-specific sub layer). The QoS can be changed dynamically and it supports extremely well data bursts. Everything is negotiated separately for uplink and downlink.

2.1 Physical Layer Details

2.1.1 OFDM Basics

OFDM is a multicarrier modulation scheme that is based on the idea of dividing a given high-bit-rate data stream into several parallel lower bit-rate streams. Each of those streams is then modulated on separate carriers, the so called subcarriers or tones. Multicarrier modulation schemes eliminate or minimize ISI by making the symbol time large enough so that the channel induced delays are an insignificant proportion of the symbol duration. For that reason, in high-data-rate systems, in which the symbol duration is relatively small and it is inversely proportional to the data rate, splitting the data stream into many parallel streams increases the symbol duration of each stream. By doing so, the delay spread is only a small fraction of the symbol duration.

OFDM is a spectrally efficient version of multicarrier modulation, in which the subscribers are selected in such way that they are orthogonal to one another over the symbol duration. In that way, it avoids the necessity to have no overlapping subcarrier channels to eliminate ISI. The first subcarrier has to have a frequency such that it has an integer number of cycles in a symbol period. The spacing between adjacent subcarriers (subscriber bandwidth - Bsc) has to be defined by the following Equation 3:

Bsc=BL (3)

Where B is the nominal bandwidth, it is also equal to the data rate. L is the number of subcarriers. This relation ensures that all tones are orthogonal to one another over the symbol period. It can be shown that the OFDM signal is equivalent to the inverse discrete Fourier transform of the data sequence block taken L at a time. Therefore, it is very straightforward to implement OFDM transmitters and receivers in discrete time using IFFT and FFT, respectively.

In order to completely eliminate the ISI, OFDM introduces guard intervals between the symbols. For the successful elimination, the guard interval should be larger than the expected multipath delay spread. On the other hand, the introduction of the guard interval implies power wastage and decrease in bandwidth efficiency. The amount of power waste depends on the proportion between the OFDM symbol duration and the guard time. Thus, the larger the symbol period, the smaller the power loss and bandwidth efficiency. Large symbol periods also mean more subcarriers.

The size of the FFT in an OFDM design should be chosen carefully as a balance between protection against multipath, Doppler shift, and design cost/complexity. For a given bandwidth, selecting a large FFT size would reduce the subcarrier spacing and increase the symbol time. This is the reason why it makes it easier to protect against multipath delay spread. In contrary, the reduced subcarrier spacing makes the system more vulnerable to intercarrier interference owing to Doppler spread in mobile applications. Careful balancing is required to compete the influence of delay and Doppler spread.

2.1.2 Advantages and Disadvantages of OFDM

There are several advantages of OFDM over other solutions for high-speed transmission.

Reduced computational complexity: OFDM can be easily implemented using FFT/IFFT, where the processing requirements grow only slightly faster than linearly with data rate or bandwidth. The computational complexity of OFDM can be represented the following Equation 4.

O(BlogBTm) (4)

Where B is the bandwidth and Tm is the delay spread. This complexity is much lower than that of standard equalizer-based system, which complexity is shown by Equation 5:

O(B2Tm) (5)

Graceful degradation of performance under excess delay: the performance of an OFDM system degrades gracefully as the delay spread exceeds the value designed for. Greater coding and low constellation sizes can be used to provide fallback rates that are significantly more robust against delay spread. In other words, OFDM is well suited for adaptive modulation and coding, which allows the system to make the best of the available channel conditions. This is in contrast with the abrupt degradation owing to error propagation that single-carrier systems experience as the delay spread exceeds the value for which the equalizer is designed.

  • Exploitation of frequency diversity : OFDM facilitates coding and interleaving across subcarriers in the frequency domain, which can provide robustness against burst errors caused by portions of the transmitted spectrum undergoing deep fades. WiMAX defines subcarrier permutations that allow systems to exploit this.
  • Use as a multi-access scheme : OFDM can be used as multi-access scheme, where different tones are partitioned among multiple users. This scheme is also known as OFDMA and is exploited in mobile WiMAX. The ability to provide fine granularity in channel allocation is also offered by the OFDMA. In relatively slow time-varying channels, it is possible to significantly enhance the capacity by adapting the data rate per subscriber according to the SNR of that particular subcarrier.
  • Robust against narrowband interference< : OFDM is relatively robust against narrowband interference, since such interference affects only a fraction of the subcarriers.
  • Suitable for coherent demodulation : it is relatively easy to do pilot-based channel estimation in OFDM system, which renders them suitable for coherent demodulation schemes that are more power efficient.

Apart from the advantages that were already described, there are two main disadvantages:

  • There is a problem associated with OFDM signals having a high peak-to-average ratio that causes nonlinearities and clipping distortion. This can lead to power inefficiencies that need to be countered.
  • The second disadvantage is related to OFDM signals which are very susceptible to phase noise and frequency dispersion, and the design must mitigate these imperfections. This also makes it critical to have accurate frequency synchronization.

2.1.3 OFDMA: Sub channelization

Sub channels are formed by the division of available subcarriers into groups. [3new] Fixed WiMAX based on the OFDM-PHY allows a limited form of sub channelization in the uplink only. There are 16 defined sub channels, where 1,2,4,8 or all sets can be assigned to a subscriber station in the uplink. Uplink sub channelization in fixed WiMAX allows SSs to transmit using only a fraction (as low as 1/16) of the bandwidth allocated to it by the BS. The scheme used allows link budget improvements and can be used to improve range performance and/or improve battery life of SSs. A typical 1/16 sub channelization factor provides a 12 dB link budget enhancement.

In contrast, mobile WiMAX uses OFDMA-PHY and it allows sub channelization in both the uplink and downlink. Sub channels effectively form the minimum frequency resource-unit, which is allocated by the BS. Therefore, different users may be allocated different sub channels as this is multiple-access mechanism. It is also known as orthogonal frequency division multiple access, which gives mobile WiMAX PHY its name.

Sub channels may be formed using wither contiguous subcarriers or subcarriers pseudo-randomly distributed across the frequency spectrum. More frequency diversity is achieved by using distributed sub carriers, which is particularly useful for mobile applications. WiMAX defines several sub channelization schemes based on distributed carriers for both the uplink and downlink. One of them is partial usage of subcarriers, which is mandatory for all mobile WiMAX implementations.

The second sub channelization scheme is based on contiguous subcarriers and is also known as adaptive modulation and coding. Although the frequency diversity is lost, band AMC allows system designers to exploit multiuser diversity, allocating sub channels to users based on their frequency response. Multiuser diversity can provide significant gains in overall system capacity, if the system strives to provide each user with a sub channel that maximizes its received SINR. Generally, the contiguous sub channels are more suited for fixed and low-mobility applications.

Fig. 9 shows a typical sub carrier structure.

FIGURE 9 + explanation

2.1.4 Frame Structure

Mobile WiMAX used to support Time Division Duplex only but recently full and half-duplex Frequency Division Duplex support has been added. [4 new] It is mainly because of local restrictions in some areas. A major disadvantage of the TDD is that it needs to be synchronised over the whole system. On the other hand, there are several reasons why TDD usage is preferred. One of them is that the ratio of DL/UL data rates can be adjusted freely. It is in contrast with the FDD where the ratio is always constant and in most cases symmetric. Channel reciprocity is assured by using TDD, which gives better support of link adaptation, MIMO and other closed loop advanced antenna technologies. Whereas FDD requires a pair channel, but TDD can share one for both DL/UL traffic. Additionally, from economical point of view, FDD transceivers are more complex and therefore more expensive to manufacture.

For the simulation project that has been set up, FDD mode was used and for this purpose the next part of this chapter explains the FDD frame structure in relation to relaying. For information purposes TDD frame structure is also discussed. Frame Format for FDD

The FDD mode does not provide guard times and parallel transmissions in DL and UL. [5new]

The BS always sets up the master frame, in other words, it generates the timing schedule for the next period. The process begins with synchronisation pilots, a broadcast channel and a random/contention access channel; a few regular frames follow, but this very much depends on the technology used. Some of the frames are reserved for the second hop(s) by the BS, which are later used in the responsibility of the relay(s). A term called stealth relaying is used when the BS does not even distinguish between relays and ordinary SSs. It is preferable to use centrally and relay-aware BS controlled resource allocation for further systems. Interleaved multihop frames are incorporated by modern multihop systems, where frames for the first and second hops alternate in time and they are all controlled by the BS.

The smallest granularity resource unit is a chunk, 100 chunks form a basic frame. Symbols, on the other hand, are used for signalling and synchronization, but they are out of focus here. They are shown in grey in Fig. 10(a) and the only impact here is the overhead-resources not available for data throughput. DL and UL transmissions happen simultaneously in the same frame raster. There are 3 ways of integrating relaying (the frames used for the second hop or beyond:

  • Time Domain Relaying: resources for hop1 and hop2 are separated in time (sequentially)
  • Frequency Domain Relaying: recourses for hop1 and hop2 are separated in frequency (neighbour band)
  • OFDMA Domain Relaying: recourses for hop1 and hop2 are separated in frequency (sub channels)

As OFDMA was already introduces as topic earlier, it is important to describe what its advantages are over OFDM relaying. First one is that the resources can be subdivided in a finer granularity than it is possible using OFDM only. Fig. 10(b) shows that first-hop transmissions are always treated the same way. They only occupy the required resources for their traffic and there is not waste due to completely assigned but incompletely filled frames. In the UL several SSs share the full bandwidth and each of them transmits on a subset of sub channels, with a guard band between them. The BS or RN coordinates the orthogonal interference-free use of these sub channels by the SSs. OFDMA subdivision in the DL is also available, where the BS and RN send on distinct sub channels, but using sufficient guard band. Even if the side band power is below the signal level, this can cause serious interference trouble at the SS when receiving the useful signal from a far distant BS and the interference from the RN nearby. In this case, interference mitigation strategies in the BS are helpful. This also leads to proper association decisions for intra-cell handover. The handover concept is introduced later in the report and the fundamental principles are discussed in details. TDD Frame Structure

Mobile WiMAX TDD mode allows the DL and UL transmissions to share the same transmission medium. [6new] Fig. 11 shows the typical structure of TDD frame. Transmission and reception cannot occur simultaneously, due to the prohibitively complex filtering, which is essential for the separation of UL and DL.

Something more, in order to avoid interference between the signals transmitted from the BS and other SSs, time gaps such as the transmit/receive transition gap or receive/transmit transition gap are needed. The TTG allows sufficient time for the BS to switch from transmit to receive mode. RTG provides time for the BS to switch from receiving to transmitting. Parameters that form the size of the time gaps are: cell radius, the transceiver turnaround time and other implementation constraints.

2.1.5 Adaptive Modulation and Coding in WiMAX

It is inevitable to mention the modulation and coding techniques that are employed by the WiMAX standard. Their principles are important as later in the report; those modulations are used for the simulation purposes and are base on the discussion how the SSs can be connected to the BSs and what effect the modulation and coding have on the connectivity.

WiMAX supports a variety of modulation and coding schemes. [3 new]This allows the scheme to change on a burst-by-burst basis per link, depending on channel conditions. The SS can provide feedback on the DL channel quality to the BS by using channel quality feedback indicator. For the UL, the BS can easily estimate the channel quality, based on the received signal quality. The BS scheduler can take into account the channel quality for each user's UL and DL and assign a modulation and coding technique that maximizes the throughput for the available SNR. Adaptive modulation and coding increases the overall system capacity, as it allows real-time trade-off between throughput and robustness on each link.

Table 2 lists various modulation and coding scheme used by WiMAX. In the DL, QPSK, 16 QAM, and 64 QAM are mandatory for both fixed and mobile WiMAX. 64 QAM is optional in the UL. FEC coding using conventional codes is mandatory as well. Conventional codes are combined with an outer Reed-Solomon code in the downlink for OFDM-PHY. Turbo codes and low-density parity check codes are optionally supported at a variety of code rates as well. There are 52 combinations of modulation and coding techniques that are defined in WiMAX as burst profiles.




BPSK, QPSK, 16 QAM, 64 QAM; BPSK optional for OFDMA-PHY

BPSK, QPSK, 16 QAM; 64 QAM optional


Mandatory: convolution codes at rate ½, 2/3, ¾, 5/6

Mandatory: convolution codes at rate ½, 2/3, ¾, 5/6

Optional: convolution turbo codes at rate ½, 2/3, ¾, 5/6; repetition codes at rate ½,1/3, 1/6, LDPC, RS-Codes for OFDM-PHY

Optional: convolution turbo codes at rate 1/2., 2/3, ¾, 5/6; repetition codes at rate ½, 1/3, 1/6, LDPC

Table 2. Modulation and Coding supported by WiMAX.

2.2 MAC Layer Details

The second important layer of the WiMAX standard is the MAC - medium access control. In general, the 802.16 MAC is designed to support multipoint-to-multipoint architecture, where a centrally located BS handles multiple independent sectors simultaneously. [7 new] On the downlink, data to SSs are multiplexed in TDM fashion. On the contrary, the uplink is shared between SSs in TDMA fashion.

IEEE standard 802.16 defines 2 general service-specific convergence sub-layers for mapping services to and from 802.16 MAC connections. The first one is the ATM convergence sub-layer, which is defined for ATM services. The second one is the packet convergence layer, defined for mapping packet services such as IPv4, IPv6, Ethernet, and virtual local area network. There are 3 primary tasks of the sub-layer:

  • To classify service data units to the correct MAC connection
  • Preserve or enable QoS
  • Enable bandwidth allocation

Depending on the type of service, the mapping takes a range of forms. Additionally to the basic convergence sub-layer tasks, the layers can also perform more complicated functions - payload header suppression and reconstruction to enhance air link efficiency.

As it was already mentioned, the 802.16 MAC is connection-oriented and all services, including inherently connectionless services, are mapped to a connection. Thus, it provides a mechanism for requesting bandwidth, associating QoS and traffic parameters, transporting and routing data to the appropriate convergence sub-layer; as well as all other actions associated with the contractual terms of service. Connections are referenced with 16-bit connection identifiers and may require continuously granted bandwidth or bandwidth on demand.

Every SS has a standard 48-bit MAC address, which serves mainly as an equipment identifier. The primary addresses used during operation are the CIDs. Three management connections are assigned to the SS when entered the network. These connections reflect the three different management connections in each direction:

  • Basic connection- used for the transfer of short, time-critical MAC and radio link control massages.
  • Primary management connection- used to transfer short longer, more delay-tolerated messages (e.g. those used for authentication and connection setup).
  • Secondary management connection- used for the transfer of standards-based management messages, such as DHCP, TFTP, and SNMP.

Apart from the management connections, transport connections for the contracted services are also allocated to the SSs. Transport connections are unidirectional and facilitate different DL and UL QoS and traffic parameters; they are typically assigned to services in pairs.

Additional connections for other purposes may also be reserved by MAC:

  • Contention-based initial access
  • Broadcast transmissions in the DL, as well as for signalling broadcast connection-based polling of SS bandwidth needs
  • Multicast, rather than broadcast, connection-based polling. SSs may be instructed to join multicast polling groups associated with these multicast polling connections.

2.2.1 MAC PDU Construction and Transmission

As it was already discussed, the MAC sub-layer is independent of the higher-layer protocol and performs such operations as scheduling, [3 new] ARQ, bandwidth allocations, modulation and code rate selection. The SDUs arriving from the higher-layer to the sub-layer are assembled to create the MAC PDU, which is the basic payload unit handled by the MAC and PHY layers. Multiple SDUs can be carried on a single MAC PDU, or a single SDU can be fragmented to be carried over multiple MAC PDUs; this is entirely based on the size of the payload. When an SDU is fragmented, the position of each fragment within the SDU is labelled by a sequence number. The usage of the sequence number is to ease assembling of the SDU from its fragments in the correct order at the receiver MAC layer.

Multiple MAC PDUs destined to the same receiver can be concatenated and carried over a single transmission opportunity or data region. This is purely done to use the PHY resources more efficiently. This is shown in Fig. 12. For non-ARQ-enabled connections, each fragment of the SDU is transmitted in sequence. For ARQ-enabled connections, the SDU is first divided into fixed length ARQ blocks. Each ARQ block has its own block sequence number. ARQ BLOCK-SIZE parameter is used by the BS for each CID to specify the length of the block. The length of the SDU should be integral multiple of the ARQ BLOCK-SIZE, if it is not - the last ARQ block is padded. Once the SDU is portioned into ARQ blocks, the partitioning remains until all the ARQ blocks have been received and acknowledged by the receiver. After the partitioning, the SDU is assembled into MAC PDUs in a normal fashion as it is shown in Fig. 12. For ARQ-enabled connections, the fragmentation and packing sub header contains the BSN of the first ARQ block following the sub header. The ARQ feedback from the receiver comes in the form of ACK indicating successful reception of the ARQ blocks. The feedback is sent either as a stand-alone MAC PDU or piggybacked on the payload of a regular MAC PDU. ARQ feedback can be two types:

  • Selective:indicates that the ARQ block has been received without errors
  • Cumulative:indicates that all blocks with sequence number less than or equal to the BSN have been received without error.

Each MAC PDU consists of a header followed by a payload and cyclic redundancy check. The CRC is based on IEEE 802.3 and is calculated on the entire MAC PDU; the header and the payload. Two types of PDUs (Fig. 13) are introduced, and each of them has different header structure:

  • Generic PDU- used to carry data and MAC-layer signalling massages. It starts with a generic header, which structure is shown on Fig.13, and then followed by a payload and a CRC. The different information elements in the header of a generic MAC PDU are shown in Table 3.
  • Bandwidth PDU- used by the SS to inform the BS that more bandwidth is needed in the UL, due to pending data transmission. This type of PDU consists of bandwidth-request header, and there is no payload or CRC. The header information elements are shown in Table 4.


Length (bits)




Header type (set to 0 for such header)



Encryption control (0= payload not encrypted; 1= payload encrypted)






Extended sub header field (1 = ES present; 0 = ES not present)



CRC indicator (1 = CRC included; 0 = CRC not included)



Encryption key sequence (index of the traffic encryption key and the initialization vector used to encrypt the payload)






Length of MAC PDU in bytes, including the header



Connection identifier on which the payload is to be sent



Header check sequence

Table 3, Generic MAC header fields


Length (bits)




Header type (set to 1 for such header)



Encryption control (set to 0 for such header)






bandwidth request (the number of bytes for uplink bandwidth request by the SS for the given CID)



Connection identifier



Header check sequence

Table 4. Bandwidth request MAC header fields

Once a MAC PDU is constructed, it is handed over to the scheduler, which schedules the MAC PDU over the PHY resources available. The scheduler checks the service flow ID and the CID of the MAC PDU, which allows it to gauge its QoS requirements. Based on the QoS requirements of the MAC PDUs belonging to different CIDs and service flow Ids, the scheduler determines the optimum PHY resource allocation for all the MAC PDUs, on a frame-by-frame basis.

2.2.2 DL and UL MAP

WiMAX MAC supports both TDD and FDD modes. [7 new] In FDD, both continuous and burst downlinks are supported.

  • Continuous downlinks allow for certain robustness improvement techniques - for example interleaving.
  • Burst downlinks allow the use of more complex robustness and capacity enhancement techniques, e.g. subscriber level adaptive burst profiling and advanced antenna systems.

The MAC builds the DL sub frame, starting with a frame control section. The frame control section contains the DL-MAP and UL-MAP messages. They indicate PHY transitions on the downlink; bandwidth allocations and burst profiles on the uplink.

The DL-MAP is applicable to the current frame and is always at least 2 FEC blocks long. In both TDD and FDD modes, the UL-MAP provides allocations starting no later than the next downlink frame. If the processing times and the round trip delays are observed, then the UL-MAP may allocate starting in the current frame. The minimum time between receipt and applicability of the UL-Map for an FDD system is shown in Fig. 14

2.2.3 Authentication, Registration and Security

Another important factor of the MAC protocol is that it manages procedures such as authentication, registration and security.

Each of the SS holds two certificates: factory installed X.509 digital certificate and certificate of manufacturer. Those certificates are sent from the SS to the BS in the Authorization Request and Authentication Information messages. The certificates establish a link between the 48 bit MAC address of the SS and its public RSA key. The network is able to identify the SS using the certificates, and afterwards it can check the authorization of the SS. If the SS is authorized to join the network, BS will respond with an Authorization Reply containing an Authorization Key. The Authorization Key is encrypted with the SS's public key and used to secure further transactions.

When the authorization is successful, the SS will be registered with the network. This will establish the secondary management connection of the SS and determine capabilities related to connection setup and MAC operation. The version of IP used on the secondary management connection is also determined during registration.

WiMAX's privacy protocol is based on the Privacy Key Management protocol. It is based on the DOCSIS BPI+ specification, but has been modified to fit seamlessly into the MAC protocol. It accommodates stronger cryptographic methods, such as the Advanced Encryption Standard.

PKM is build around the concept of security associations. SA is a set of cryptographic methods and the associated keying material. It contains the information about which algorithms to apply, which key to use, and so on. Every SS establishes at least one SA during initialization. Apart from the basic and primary management connections, all other connections are mapped to an SA either at connection setup time or dynamically during operation.

For the traffic encryption, the PKM uses the Data Encryption Standard, running in the cipher block chaining more with 56-bit keys. The CBC initialization vector depends on the frame counter and differs from the frame to frame. To reduce the number of computationally intensive public key operations during normal operation, the transmission encryption keys are exchanged using 3DES with a key exchange key derived from the authorization key.

The PKM protocol messages are authenticated using the Hashed Message Authentication Code protocol with SHA-1. Message authentication in essential MAC functions, such as the connection setup, is provided by the PKM protocol.

2.2.4 Services and Parameters

The last part of the MAC layer is the services and their parameters. It is an essential part of the standard implementation and this part of the chapter lists all QoS that are provided by the WiMAX family.

Scheduler is the one that controls the scheduling services for the standard and handles the mechanisms supported by the MAC to transport data. [9 new] Each connection is associated with a single data service. Data services are associated with a set of QoS parameters that define their behaviour. Two data parameters are used for the management of the parameters: dynamics service addition and dynamic service change. Five different services are supported: unsolicited grant services, real-time polling services, extended real-time polling services non-real-time polling services, and best effort. Table 5 lists the application and specification for each of those services. [10 new]

QoS Category


QoS Specifications



Maximum sustained rate

Maximum latency tolerance

Jitter tolerance


Streaming audio and video

Minimum reserved rate

Maximum sustained rate

Maximum latency

Traffic priority


Voice with activity detection (VoIP)

Minimum reserved rate

Maximum sustained rate

Maximum latency tolerance

Jitter tolerance

Traffic priority


File transfer protocol

Minimum reserved rate

Maximum sustained rate

Traffic priority


Data transfer, web browsing, etc.

Maximum sustained rate

Traffic priority

Table 5. Mobile WiMAX applications and QoS

  • UGS- supports real-time data streams. Each stream is of fixed-size data packets issued at periodic time intervals. It is particularly useful for applications such as VoIP without silence suspension. Mandatory service flow parameters are: maximum sustained rate, maximum latency tolerance, jitter tolerance and request/transmission policy. If the policy is present, the minimum reserved traffic rate parameter will be the same value as the maximum sustained traffic rate parameter.
  • rtPS- supports real-time data streams, each of which is issued at periodic intervals. It is useful for MPEG videos and consists of variable-size data packets. Compulsory parameters are: minimum reserved rate, maximum sustained rate, maximum latency, traffic priority, and again request/traffic policy.
  • ertPS- it is a combination of USG and rtPS. [4new] ertPS is designed to support VoIP with activity detection, which means that when there is a silence period in the communication, the bandwidth can be saved by downsizing the packets length. The maximum sustained traffic rate, which defines the default size of the allocations, may be changed with ertPS.
  • nrtPS- used for delay-tolerant data streams, which consist of variable data packets. Minimum data rate is required as it used the FTP. Flow parameters are: minimum reserved rate, maximum sustained rate, traffic priority, and request/transmission policy.
  • BE - is used to support data streams for which no minimum service level is required and therefore may be handled on a space-available basis. Associated parameters are: maximum sustained rate, traffic priority, and request/transmission policy.

3 Handovers

The aim of the next chapter is to introduce the term handovers and to explain their importance for the WiMAX standard features.

Handovers are very vital part of a wireless technology.[11new] When an SS moves between different BSs, the connection should also move and in order to do so a seamless handover should be performed. By the term seamlessness, it defines the necessity to maintain current session, QoS and service level agreements during and after the handover. In other words, the seamless handover should not be noticeable by the user; however, this very much depends on the kind of services the user is requiring. With real-time applications, such as videoconferencing or streaming media, s slight decrease of the connection may be observed. In contrast, while a user is browsing a website or transfers files, they will most probably not notice any change in their connection. There are two crucial factors related to the handover procedure: the latency and the packet loss. They have to be as small as possible to make the handover seamless. Several reasons why and when a handover should be initiated are listed below:

  • SS current position and velocity - high velocity may lead to different handover decisions
  • Link quality - in case a BS is overloaded, the network can decide to relocate some of the SSs
  • Conserving battery power - for battery saving purposes, an SS may chose to switch to a closer station and be energy efficient
  • Context and requirements - if an SS requires different type of service, it may be necessary to change BS.
  • Handovers have a significant effect on the performance of channel allocation algorithms. At high traffic loads, the majority of forced call terminations are as a result of the lack of channels available for handover rather than to interference. This is a major problem in microcellular systems, where the rate of handovers is considerably higher than that in normal cellular systems.
  • There are a number of solutions to reduce the performance penalty caused by handovers. One of them is to reserve some channels solely for handovers, generally referred to as cut-off priority or guard channel schemes. However, this solution reduces the maximum amount of carried traffic or system capacity and hence yields increased new call blocking.
  • Algorithms that give higher priority to requests for handovers than to new calls are called Handover prioritization schemes. Guard channel schemes are therefore a type of handover prioritization arrangement. Another type of handover prioritization is constituted by handover queuing schemes. Normally, when an allocation request for handoff is rejected, the call is forcibly terminated. By allowing handover allocation requests to be queued temporarily, the forced termination probability can be reduced. The simplest handover queuing schemes use a First-In First-Out (FIFO) queuing regime. A non-pre-emptive priority handover queuing scheme in which handover requests in the queue that are the most urgent ones are served first.
  • A further alternative to help reduce the probability of handover failure is to allow allocation requests for new calls to be queued. New call allocation requests can be queued more readily than handovers because they are less sensitive to delay. Handover queuing reduces the forced termination probability owing to handover failures but increase the new call blocking probability. New call queuing reduces the new call blocking probability and also increases the carried teletraffic. This is because the new calls are not immediately blocked but queued, and in most cases they receive an allocation later.

Handovers have a significant effect on the performance of channel allocation algorithms. At high traffic loads, the majority of forced call terminations are as a result of the lack of channels available for handover rather than to interference. This is a major problem in microcellular systems, where the rate of handovers is considerably higher than that in normal cellular systems.

There are a number of solutions to reduce the performance penalty caused by handovers. One of them is to reserve some channels solely for handovers, generally referred to as cut-off priority or guard channel schemes. However, this solution reduces the maximum amount of carried traffic or system capacity and hence yields increased new call blocking.

Algorithms that give higher priority to requests for handovers than to new calls are called Handover prioritization schemes. Guard channel schemes are therefore a type of handover prioritization arrangement. Another type of handover prioritization is constituted by handover queuing schemes. Normally, when an allocation request for handoff is rejected, the call is forcibly terminated. By allowing handover allocation requests to be queued temporarily, the forced termination probability can be reduced. The simplest handover queuing schemes use a First-In First-Out (FIFO) queuing regime. A non-pre-emptive priority handover queuing scheme in which handover requests in the queue that are the most urgent ones are served first.

A further alternative to help reduce the probability of handover failure is to allow allocation requests for new calls to be queued. New call allocation requests can be queued more readily than handovers because they are less sensitive to delay. Handover queuing reduces the forced termination probability owing to handover failures but increase the new call blocking probability. New call queuing reduces the new call blocking probability and also increases the carried teletraffic. This is because the new calls are not immediately blocked but queued, and in most cases they receive an allocation later.

There are two main types of handovers: horizontal (handovers within the same technology) and vertical (handovers between different network access technologies).

  • Horizontal handovers- are Layer-2 handovers and are also referred to as “micro-mobility”. In this case only the BS is changed and the IP-information is kept the unchanged. This kind of handover causes small latency and low packet loss
  • Vertical handovers- Layer-3 handovers and are also referred as “macro-mobility”. It changes the IP-connection point, and as a result the IP information is changed as well. As a result, significant latency and packet loss are observed.

3.1 Handover Types

WiMAX specification supports three types of handovers: Hard Handover, Fast Base Station Switching, and Macro Diversity Handover.[4new] HHO is compulsory, whereas the other two are optional. The WiMAX forum has been working on the enhancement of the HHO techniques to achieve handovers in less than 50 milliseconds.

3.1.1 Hard Handover

HHO is a procedure that changes the serving BS using a “break-before-make” approach. Effectively, this means that the connection to the old BS is broken before the new BS is connected to the SS. Using this approach, excess signalling traffic is avoided during the handover procedure, however, the time before the connection is again in normal operation may be longer. While the SS is connected to a BS, it listens to the link-layer messages for another BS, which periodically broadcasts neighbour advertisement messages. The advertising messages are used for network identification and recognition of the services that are provided. Facts about the signal quality from a neighbouring BS are included in the messages. In case a better BS is not found, the SS stores the information that has been received for future handovers. Fig. 15 illustrates a typical scenario of HHO, when a moving user reaches the point where the signal level is of better quality with another BS. A further decision criterion has to be set up to avoid constant handover back and forth between BSs.

3.1.2 Macro Diversity Handover

When MDHO is supported by MS and by BS, the “Diversity Set” (or in some cases it is called “Active Set”) is maintained by SS and BS. Diversity set is a list of the BS's, which are involved in the handover procedure. Diversity set is defined for each of SS's in network. SS communicates with all BS's in the diversity set (Fig. 16).

For downlink in MDHO, two or more BS's transmit data to SS such that diversity combining can be performed at the SS. For uplink in MDHO, MS transmission is received by multiple BS's where selection diversity of the received information is performed. The BS, which can receive communication among SS's and other BS's, but the level of signal strength is not sufficient is noted as “Neighbour BS”. While the SS is moving towards the neighbouring BS, at some moment the signal from the neighbouring BS becomes strong enough and the BS can be included in the active set. The factor that is used to measure the inclusion/exclusion of the BS is the long term CINR.

There are two ways that the SS uses to monitor DL control information and broadcast messages:

  • It listens only to the anchor BS for burst allocation information of other BSs in the active set
  • Listens to all BSs in the active set.

By listening to all active set BSs, a DL/UL-MAP message from any BS may include information for other BSs. The MDHO is started when the SS decides to receive/transmit from multiple BSs at the same time. For DL traffic, two or more BSs transmit the data to the SS, which performs diversity combining. For the UL traffic, the traffic generated by the SS is received by all BSs in the active set and selection diversity is then performed.

Interestingly, the MDHO requires all BSs to communicate on synchronised basis, because of the frames sent by the BSs at a certain time frame have to be received by the SS within the specific prefix interval. The BSs frame structure have to be synchronised and the frequency assignments to have the same values. Furthermore, the same set of CIDs and MAC/PHY PDUs sent to the SS have to be generated by the BSs. The encryption information and network entry exchanged information have to be shared between the SS and the BSs as well.

3.1.3 Fast Base Station Switching

In FBSS, the SS and BS diversity set is maintained similar as in MDHO. SS continuously monitors the base stations in the diversity set and defines an “Anchor BS”. Anchor BS is only one base station of the diversity set that MS communicates with for all uplink and downlink traffic including management messages (Fig. 17). This is the BS where SS is registered, synchronized, performs ranging and there is monitored downlink channel for control information. The anchor BS can be changed from frame to frame depending on BS selection method, which means each frame can be sent via different BS in diversity set.

Generally, the requirements for the FBSS are the same as the ones of MDHO without the demand for the same CIDs and MAC/PHY PDUs.