WASHINGTON UNIVERSITY SEVER INSTITUTE SCHOOL OF ENGINEERING AND APPLIED SCIENCE DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

Transcription

1 WASHINGTON UNIVERSITY SEVER INSTITUTE SCHOOL OF ENGINEERING AND APPLIED SCIENCE DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING EXPONENTIAL EFFECTIVE SIGNAL TO NOISE RATIO MAPPING (EESM) COMPUTATION FOR WIMAX PHYSICAL LAYER by Abdel Karim Al Tamimi Prepared under the direction of Professor Raj Jain A thesis presented to the Sever Institute of Washington University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE May 2007 Saint Louis, Missouri

2 WASHINGTON UNIVERSITY SEVER INSTITUTE SCHOOL OF ENGINEERING AND APPLIED SCIENCE DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING ABSTRACT EXPONENTIAL EFFECTIVE SIGNAL TO NOISE RATIO MAPPING (EESM) COMPUTATION FOR WIMAX PHYSICAL LAYER by Abdel Karim Al Tamimi ADVISOR: Professor Raj Jain May 2007 St. Louis, Missouri WiMAX (IEEE ) is being introduced as one of the major future key technologies for wireless broadband. Performance modeling and simulations are required to obtain the best performance from WiMAX deployments. In WiMAX, the channel is divided into thousands of orthogonal subcarriers resulting is what is called Orthogonal Frequency Division Multiple Access (OFDMA). One of the major challenges in modeling WiMAX is to evaluate the channel quality and model the combined effect of interference on these subcarriers. EESM (Effective Exponential SINR Mapping) is a commonly used method to combine signal to interference and noise ratios (SINR) in such multi-carrier environments. EESM requires the use of a "beta" parameter that needs to be set correctly so that the channel model results in accurate block error rate (BLER) for a given modulation and coding scheme (MCS). In this thesis a simulation model of the WiMAX physical layer is implemented and several experiments are conducted to determine the beta values for a number of MCS used in WiMAX networks.

3 To my advisor, my family and my friends ii

4 Contents List of Tables... v List of Figures... vi Acknowledgements... vii Acronyms... viii Glossary... xi 1 WiMAX Introduction WiMAX Concepts WiMAX Protocol Layers WiMAX Compared to 3G and WiFi WiMAX Standards WiMAX Physical Layer Introduction Basic Concepts and Definitions OFDM/OFDMA in WiMAX OFDM Basics OFDMA Basics OFDM in WiMAX OFDMA Subchannelization PHY Frame Structure TDD Frame Components WiMAX Physical Layer Components Permutation Schemes EESM Introduction EESM Simulation Setup and Results Introduction Downlink Model Downlink Transmitter Channel Model Downlink Receiver and BLER Block Simulation Setup...33 iii

5 4.4 Beta Calibration Steps Results Conclusion...44 References...45 Vita...47 iv

6 List of Tables Table 1.1: WiMAX Standards...6 Table 2.1: Subcarrier Distribution for FFT Table 2.2: QPSK bits to Symbol Mapping...20 Table 2.3: Parameters of PUSC Permutation in OFDMA-DL...21 Table 3.1: Beta Values from 3GPP...27 Table 3.2: Beta Values Using Ped A Channel Model...27 Table 3.3: Beta Values Using Ped B Channel Model...27 Table 3.4: Delay and Power value for Ped A and Ped B Channel Models...28 Table 3.5: Delay and Power value for Veh A and Veh B Channel Models...28 Table 4.1: Simulation Parameters...33 v

7 List of Figures Figure 1.1: WiMAX World...2 Figure 1.2: WiMAX Architecture...3 Figure 1.3: WiMAX Protocol Layers...4 Figure 2.1a: TDD...11 Figure 2.1b: FDD...11 Figure 2.2: Comparison between FDM and OFDM Efficiency...12 Figure 2.3: OFDMA Frequency Allocation...13 Figure 2.4: Subcarriers Distribution...14 Figure 2.5: OFDMA WiMAX Frame in TDD Mode...16 Figure 2.6: WiMAX Physical Layer Conceptual Diagram...18 Figure 2.7: QPSK Modulation Constellation...20 Figure 2.8: Square-root Filter Equation...20 Figure 2.9: DL-PUSC Subcarrier Permutation Scheme...22 Figure 2.10: WiMAX System Using AMC Permutation...23 Figure 3.1: EESM Functionality...25 Figure 3.2: EESM Formula...26 Figure 3.3: Beta Calibration Formula...26 Figure 4.1: Downlink Simulation Model...30 Figure 4.2: Downlink Transmitter Model...30 Figure 4.3: Channel Model Block...32 Figure 4.4: Downlink Receiver Block...33 Figure 4.5: AWGN Reference Curve for QPSK ½...35 Figure 4.6: Eb/No vs BLER for QPSK 1/ Figure 4.7: Eb/No vs BLER for QPSK 3/ Figure 4.8: Eb/No vs BLER for 16QAM 1/ Figure 4.9: Eb/No vs BLER for 16QAM 3/ Figure 4.10: Eb/No vs BLER for 64QAM 2/ Figure 4.11: Eb/No vs BLER for 64QAM 3/ Figure 4.12: SNR vs BLER for QPSK 1/ Figure 4.13: SNR vs BLER for QPSK 3/ Figure 4.14: SNR vs BLER for 16QAM 1/ Figure 4.15: SNR vs BLER for 16QAM 3/ Figure 4.16: SNR vs BLER for 64QAM 2/ Figure 4.17: SNR vs BLER for 64QAM 3/ Figure 4.18: Eb/No vs BLER...42 Figure 4.19: SNR vs BLER...42 vi

8 Acknowledgements Thanks to my advisor and my professor who gave a lot of his time to help me. Special thanks to Dr. Vafa Ghazi-moghadam from CoWare, Dr. Jalloul Louay from Beceem Networks, and Dr. John Kim from Sprint for their services as external advisors and helping me with their time and patience. Abdel-Karim Al Tamimi Washington University in St. Louis May 2007 vii

9 Acronyms 3G AMC ASCA AWGN BLER BPSK BS BP BTC BWA CC CDMA CID CINR CP CSMA/CA CTC DAMA DSCA EB/NO EESM FCH FDD FDMA FEC FFT FUSC IFFT ISI ISP LAN LDPCC LOS MAC MAN MCS MS OFDM OFDMA PHY Third Generation of Wireless Communication Adaptive Coding and Modulation Adjacent Subcarrier Allocation Additive White Gaussian Noise Block Error Rate Binary Phase Shift Keying Base Station Burst set Preamble Block Turbo Coding Broadband Wireless Access Convolutional Coding Code Division Multiple Access Connection ID Carrier to Interference plus Noise Ratio Cyclic Prefix Carrier Sense Multiple Access with Collision Avoidance Convolutional Turbo Coding Demand Assigned Multiple Access Distributed Subcarrier Allocation Energy per Bit (EB) to the Spectral Noise Density (NO) Ratio Exponential Effective SINR Mapping Frame Control Header Frequency Division Duplexing Frequency Division Multiple Access Forward Error Correction Fast Fourier Transform Full Usage of Subscribers Inverse Fast Fourier Transform Inter Symbol Interference Internet Service Provider Local Area Network Low Density Parity Check Coding Line of Site Media Access Control Metropolitan Area Network Modulation and Coding Scheme Mobile Station Orthogonal Frequency Division Multiplexing Orthogonal Frequency Division Multiple Access Physical Layer viii

10 PKM PMP PRBS PUSC QAM QoS QPSK RF RS-CC RTG RXDS SINR SNR SS TDD TDMA TTG WiFi WiMAX WirelessMAN Primary Key Management Point to Multipoint Pseudo Random Binary Sequence Partial Usage of Subscribers Quadtrative Amplitude Modulation Quality of Service Quadrature Phase Shifting Keying Radio Frequency Reed-Solomon concatenated with Convolutional Coding Receive-Transmit Transmission Gap Receiver Delay Spread clearing interval Signal to Interference plus Noise Ratio Signal to Noise Ratio Subscriber Station Time Division Duplexing Time Division Multiple Access Transmit-Receive Transmission Gap Wireless Fidelity Worldwide Interoperability for Microwave Access Wireless Metropolitan Area network ix

11 Glossary TERM AAS Adaptive Modulation AMC Asynchronous Attenuation AWGN Backbone BPSK Description Adaptive Antenna System: A system adaptively exploiting more than one antenna to improve the coverage and the system capacity. A system s ability to communicate with another system using multiple burst profiles and a system s ability to subsequently communicate with multiple systems using different burst profiles. Advanced Modulation and Coding. This method uses adjacent subcarriers to form subchannels. When used with fast feedback channels it can rapidly assign a modulation and coding combination per subchannel. A form of concurrent input and output communication transmission with no timing relationship between the two signals. A loss of signal strength usually related to the distance the signal must travel. Radio signal attenuation may be due to atmospheric conditions, antenna design / positioning, obstacles, etc. Attenuation is measured in decibels. Additive White Gaussian Noise, it has a uniform spectral density over a range of frequencies The core infrastructure of a networkand the central part of a network that transports information from one central location to another central location. Binary Phase shift keying is a modulation technique where the carrier is shifted by 180 degrees in accordance with a digital bit stream. 0 does not produce a phase transition where as 1 causes a phase transition to occur. x

12 Broadband A signal that involves a relatively wide range of frequencies. Often used to indicate use of a wide ("broad") range ("band") of frequencies being used in various electronics and telecommunications related subject areas. BS Base Station: A central radio transmitter/receiver used for maintaining communications with the mobile radios (or wireless-capable devices) within its range. In cellular systems, each cell (or micro-cell) has its own base station; with each base station in turn being interconnected with other cells base stations. It is generalized equipment set providing connectivity, management, and control of the subscriber station (SS). BSS Basic Service Set. BSS usually comprises of an access point (an AP) and all the LAN PCs (or other wireless capable client devices) that may be associated with it Burst A burst contains payload data and is formed according to the rules specified by the burst profile associated with the burst. A burst is a complete unit of transmission that includes a leading preamble, encoded payload, and trailing termination sequence. Burst Frame A burst frame contains all information included in a single transmission. It consists of one or more burst sets. Burst Profile Set of parameters that describe the uplink or downlink transmission properties associated with an interval usage code. Each profile contains parameters such as modulation type, forward error correction (FEC) type, preamble length, guard times, etc. Burst Set A burst set is a self-contained transmission entity consisting of a preamble, one or more concatenated bursts, and a trailing termination sequence. Carrier A radio signal used to modulate the message signal. Various parameters of the carrier can be modified such as phase, xi

13 CDMA CHID CID Concatenation Convolution Codes CSMA CTS DC Subcarrier xii amplitude, frequency. Code Division Multiple Access. CDMA is a form of multiplexing -a method of multiple access that does not divide up the channel using different time-slots (as in TDMA), or using frequency modulation (such as FDMA), but encodes data by associating a special code with each channel. CHannel Identifier: An identifier used to distinguish between multiple uplink channels, all of which are associated with the same downlink Channel. Connection Identifier: It maps to a service flow identifier (SFID), which defines the Quality of Service (QoS) parameters of the service flow associated with that connection. Security associations (SAs) also exist between keying material and CIDs The act of combining multiple medium access control (MAC) protocol data units (PDUs) into a single PHY SDU (Service Data Unit). A class of codes which can detect and correct errors, where the code generated depends not only upon the present bits but also on the preceding bits in time. Carrier Sense Multiple Access - A listen before talk scheme used to mediate the access to a transmission resource. All stations are allowed to access the resource but are required to make sure the resource is free before transmitting. Clear to send is a signal from the receiving station to the transmitting station granting permission to transmit data. In IEEE a station responds to a RTS with a CTS frame, providing clearance for the requesting station to send data. In an orthogonal frequency division multiplexing (OFDM) or orthogonal frequency division multiple access (OFDMA) signal, the subcarrier whose

14 frequency would be equal to the RF center frequency of the station. DCD Downlink Channel Descriptor: A MAC message that describes the PHY characteristics of a downlink channel. DFS Dynamic Frequency Selection: The ability of a system to switch to different physical RF channels between transmit and receive activity based on channel measurement criteria. DIUC Downlink Interval Usage Code: An interval usage code specific to a downlink. DLBTG DownLink Burst Transition Gap: Gap included on the trailing edge of each allocated downlink burst so that ramp down can occur and delay spread can clear receivers. DLFP Downlink Frame Prefix contains Location and profile of the first downlink burst. The location and profile of the maximum possible number of subsequent bursts is also specified in the DLFP (DL Frame Prefix) is a data structure transmitted at the beginning of each frame and contains information regarding the current frame and is mapped to the FCH.) Downlink The direction from the base station (BS) to the subscriber station (SS). Encoder A piece of hardware or software that encodes the data,i.e., accepts the message bits and adds redundancy according to a prescribed rule there by producing encoded data at a higher bit rate. Equalization Equalization is the process to shape the received pulses so as to compensate the effects of amplitude and phase distortions caused by imperfections in the transmission characteristics of the channel. It refers to any signal processing or filtering technique used to reduce ISI. Fading The variation in received signal s amplitude FDD Frequency Division Duplex: A duplex scheme in which uplink and downlink xiii

15 transmissions use different frequencies but are typically simultaneous. Frame A structured data sequence of fixed duration used by some PHY specifications. A frame may contain both an uplink subframe and a downlink subframe. Frequency Offset Index An index number identifying a particular subcarrier in an orthogonal frequency division multiplexing (OFDM) or orthogonal frequency division multiple access (OFDMA) signal, which is related to its subcarrier index. Frequency offset indices may be positive or negative. Full Duplex A full-duplex communication system allows for simultaneous flow of information/communication in both FUSC xiv directions. In Fully Used Sub-channelization the pilot tones are allocated first and then the remaining subcarriers are divided into data subchannels.(all subchannels are allocated to the base station.) Guard Band Unused frequency spaces between channels which prevents overlapping of spectrum. HIPERLAN HIgh PErformance Radio LAN -a Wireless LAN standard. This European alternative for the American IEEE x standards has been defined by the European Telecommunications Standards Institute (ETSI) through their BRAN (Broadband Radio Access Networks) project initiative. IEEE Initial Ranging Connection Identifier released in June 2004 specifies the air interface of fixed broadband wireless access (BWA) systems supporting multimedia services. A well-known CID that is used by a subscriber station (SS) during the initial ranging process. This CID is defined as constant value within the protocol since an SS has no addressing information available until the initial ranging process is complete.

16 Interval Usage Code A code identifying a particular burst profile that can be used by a downlink or uplink transmission interval. LOS In line of sight transmission the transmitting and receiving stations (antennas) can see each other. It s a clear path between transmitting and receiving stations. MAC Medium Access Control layer is the lower layer in OSI model prior to PHY layer. The primary functions of the MAC layer are to control and access the physical medium, and also perform fragmentation and de fragmentation of packets. MAN Metropolitan Area Network. MANs are typically large (campus-wide or citywide) computer networks based on technologies such as ATM, FDDI, Ethernet (Metro Ethernet), and SMDS. MANs are often used to interlink college campus LANs, and then to further extended them by connecting them to form even bigger wide area networks or WANs. MANs capable of providing aggregated high data transfer rates (34 to 155 Mbit/s) over long (often 30 to 50 miles) distances are not atypical. Minislot A unit of uplink bandwidth allocation equivalent to n physical slots (PSs), where n = 2m and m is an integer ranging from 0 through 7. Modulation Process by which some characteristics of the message signal are varied in accordance with the modulating wave Multipath In addition to direct path from transmitter to receiver there exist several indirect paths. The interference caused due to these indirect paths is called multipath. Multiplexing Technique where multiple channels are combined for transmission over a single transmission path. NLOS In non line of sight transmission technique the stations/antennas (transmitting or receiving) need not see each other; the path is not clear which results in signal xv

17 degradation. Node A network station. A node may behave as a BS, SS, or both, and will generate and forward data to other nodes. Noise Unwanted signal superimposed on a true signal. OFDM Orthogonal Frequency Division Multiplexing is a modulation technique in which a radio signal is divided into multiple narrow frequency bands to transmit large amounts of data a, g and d use OFDM. OFDMA Orthogonal Frequency Division Multiple Access. It s a logical extension of OFDM and a modulation/multiple access technique. IEEE e uses OFDMA. Packet A unit of data. Each message sent between two network devices is often subdivided into packets by the underlying hardware and software. Depending on the protocol the packets have their own formats. PDU Protocol Data Unit: The data unit exchanged between peer entities of the same protocol layer. On the downward direction, it is the data unit generated for the next lower layer. On the upward direction, it is the data unit received from the previous lower layer PER Packet Error Rate: Packet error rate is an average fraction of transmitted packets Permutation Zone xvi that are not detected correctly. Permutation Zone is a number of contiguous OFDMA symbols, in the DL or the UL, that use the same permutation formula. The DL subframe or the UL Subframe may contain more than one permutation zone. Phase Offset Difference in reference phase of transmitted waveform and received waveform is called phase offset, expressed in degree. PHY Pilot Common IEEE abbreviation for the physical layer. A single frequency signal which is

18 transmitted for synchronization or reference purposes. POTS Plain Old Telephone Service. The term POTS usually denotes old/analog telephone devices. Preamble A preliminary signal that is transmitted to control signal detection and achieve synchronization between transmitters and receivers in wired and wireless networks. Protocol A set of rules and regulations for communication. PS Physical Slot: A unit of time, dependent on the PHY specification, for allocating bandwidth PTP Point To Point: A mode of operation whereby a link exists between two network entities. Pulse Shaping It is a process to alter the pulse shape in other words tailoring the pulse shape in a controlled manner to overcome ISI. Puncturing To generate binary code rates of 2/3, 3/4, 5/6, and 7/8, the rate 1/2 encoder outputs shall be punctured. PUSC Partially Used Sub-Channelization. The set of used subcarriers, that is, data and pilots, is first partitioned into subchannels, and then the pilot subcarriers are allocated from within each subchannel. (where some of the subchannels are allocated to the transmitter) QAM Quadrature Amplitude Modulation is a modulation technique which uses different phases such as 16, 32, 64, and 256 and each state is defined by a specific QOS xvii amplitude and phase. Quality of Service refers to the capability of a network to provide better service QPSK Quadrature Phase Shift keying: A modulation method that encodes bits as phase shifts. One of four phase shifts can be selected to encode two bits. Randomization Randomization is a process to systematically or randomly reorder (shuffle) the data. Randomization starts

19 Receiver Sensitivity RF Center Frequency RTG RTS Sampling Rate with the original data and calculates the appropriate test statistic on each reordering. The weakest signal power that can be correctly decoded. The center of the frequency band in which a base station (BS) or SS is intended to transmit. Receive/Transmit Transition Gap: A gap between the uplink burst and the subsequent downlink burst in a time division duplex (TDD) transceiver. This gap allows time for the base station (BS) to switch from receive to transmit mode and SSs to switch from transmit to receive mode. During this gap, the BS and SS are not transmitting modulated data but simply allowing the BS transmitter carrier to ramp up, the transmit/receive (Tx/Rx) antenna switch to actuate, and the SS receiver sections to activate. Not applicable for FDD systems. Request to send is a signal from the transmission station to the receiving station requesting permission to transmit data. In networks a station sends a RTS frame to another station as the first phase of a two-way handshake necessary before sending the data. xviii Sampling rate defines the no of samples taken from a continuous signal SDU Service Data Unit: The data unit exchanged between two adjacent protocol layers. On the downward direction, it is the data unit received from the previous higher layer. On the upward direction, it is the data unit sent to the next higher layer Segment A subdivision of the set of available OFDMA subchannels (that may include all available subchannels). One segment is used for deploying a single instance of the MAC. SS Subscriber Station: A generalized equipment set providing connectivity

20 between subscriber equipment and a base station (BS). SSID Service Set Identifier an alphanumeric character sequence up to 32 long uniquely identifies a group of wireless network devices used in a given "Service Set". SSRTG SS RX/TX GAP: The minimum receive to transmit turnaround gap. SSRTG is measured from the time of the last sample of the received burst to the first sample of the transmitted burst, at the antenna port of the SS. SSTTG SS TX/RX GAP: The minimum transmit to receive turnaround gap. It is measured from the time of the last sample of the transmitted burst to the first sample of the received burst, at the antenna port of the SS. Subcarrier Index An index number identifying a particular subcarrier in an orthogonal frequency division multiplexing (OFDM) or orthogonal frequency division multiple access (OFDMA) signal. Subcarrier indices are greater than or equal to zero. Synchronous A form of communication transmission with a direct timing relationship between input and output signals. The transmitter and receiver are in sync and signals are sent at a fixed rate. TCP/IP Transmission control protocol (TCP)/Internet protocol (IP). TCP guarantees delivery of data and also guarantees the order in which they were sent while IP takes care of the addressing. TDD Time Division Duplex: A duplex scheme where uplink and downlink transmissions occur at different times but may share the same frequency. TDM Burst Time Division Multiplexing Burst: A contiguous portion of a TDM data stream using PHY parameters, determined by the Downlink Interval Usage Code (DIUC) that remains constant for the duration of the burst. TDM bursts are not separated xix

21 by gaps or preambles. TDMA A method of digital wireless communications transmission allowing a large number of users to access a single radio-frequency channel without interference. Each user is given a unique time slot within each channel. TDMA Burst Time Division Multiple Access Burst: A contiguous portion of the uplink or downlink using PHY parameters, determined by the Downlink Interval Usage Code (DIUC) or Uplink Interval Usage Code (UIUC), that remains constant for the duration of the burst. TDMA bursts are separated by preambles and are separated by gaps in transmission if subsequent bursts are from different transmitters. TTG Transmit/Receive Transition Gap: A gap between the downlink burst and the subsequent uplink burst in a time division duplex (TDD) transceiver. This gap allows time for the base station (BS) to switch from transmit to receive mode and SSs to switch from receive to transmit mode. During this gap, the BS and SS are not transmitting modulated data but simply allowing the BS transmitter carrier to ramp down, the transmit/receive (Tx/Rx) antenna switch to actuate, and the BS receiver section to activate. Not applicable for FDD systems. Turbo Decoding Iterative decoding, using soft inputs and soft outputs. UCD Uplink Channel Descriptor: A medium access control message that describes the PHY characteristics of an uplink. UIUC Uplink Interval Usage Code: An interval usage code specific to an uplink. UL-MAP UpLink MAP: A set of information that defines the entire access for a scheduling interval. Uplink The direction from a subscriber station to the base station (BS). xx

22 WiMAX Wireless Access Worldwide interoperability for Microwave Access a standards-based wireless technology which provides broadband connections over long distances. End-user radio connection(s) to core networks. WirelessMAN WirelessMAN stands for: Wireless Metropolitan Area Network (i.e. Wireless MAN), and denotes use of wireless antennas to create a MAN. WirelessMANs are based on IEEE (last mile link based on GHz frequency band) specifications for the wirelessman Air Interface. WirelessMANs require line of sight, but they not only support point-tomultipoint topology, frequency-division duplex (FDD) and time-division duplex (TDD); they also provide excellent quality of service (QoS). WMANs based on the newer a make use of 2-11 GHz frequency bands, and support mesh (not just the usual point-to-multipoint) network architecture. [16] xxi

23 1 Chapter 1 WiMAX This chapter will introduce the reader to the world of wireless broadband represented by WiMAX. It illustrates WiMAX s protocol layers and architecture, and gives a brief description about the current standard. 1.1 Introduction WiMAX (Worldwide Interoperability for Microwave Access) Forum is an industry consortium that is developing interoperability requirements for equipment based on IEEE standards. The equipment following these requirements and networks constructed using such equipment are called WiMAX compliant or WiMAX. Thus, WiMAX is a point to multipoint (PMP) technology intended to provide wireless broadband solution. Based on IEEE Standard There are a number of IEE variations of IEEE specifications, e.g., IEEE , IEEE , and IEEE e WiMAX Forum is concentrating on IEEE e based networks also known as Mobile WiMAX. It can support a range of communication up to 30 miles or connection bandwidth up to 75 Megabits per second [1]. Note that the actual datarate obtainable depends upon the distance and so high data rate such as 70 Mbps is obtainable only at short distances while the data rate at long distances such as 30 miles will be an order of magnitude less. WiMAX works as a wireless Metropolitan Area Network (MAN), providing wireless broadband access especially in places where physical infrastructure is infeasible (like in rural area and third world countries). The idea is similar to a TV broadcast network

24 2 (Shown in Fig.1). Such capabilities of WiMAX support it to be a decent replacement to current cable and DSL connections [2]. Figure 1.1 WiMAX World [3] The main target of implementing WiMAX technology is to provide a convenient solution to the last mile access problem, where the high-speed data backbone traffic is to be distributed among consumers. It is expected that WiMAX will convert urban area and cities to become metro zones allowing accessibility to portable devices outdoors. 1.2 WiMAX Concepts WiMAX uses fixed wireless connection as its infrastructure, where the backbone base station is connected to the backbone network. Using the base stations scattered around the coverage area, WiMAX can offer Internet connectivity to enterprises and households.

25 3 WiMAX divides and allocates both upload (Subscriber Station (SS) to Base Station (BS)) and download (BS to SS) slots to users depending on their individual needs on a real time needed basis. Depending on the level of QoS (Quality of Service) agreement between the Internet Service Provider (ISP) and the subscriber, a guaranteed level of service can be assured. Also the ISP can provide better services to certain types of consumers. WiMAX is considered a promising solution to the last mile problem, in the same way as WiFi (Wireless Fidelity) provided a convenient solution to Local Area Networks (LAN). WiMAX has two main topologies: Point-to-Multipoint (PMP), in which it provides the connectivity needed between the BS and the subscribers, and Point-to-Point (PP) for backhaul connections between base stations. Fig. 1.2 WiMAX Architecture

26 4 Figure 2 above shows the general architecture of WiMAX. WiMAX and WiFi can be considered a complete wireless solution to Internet connection. 1.3 WiMAX Protocol Layers protocols consist of four layers as shown in Fig. 1.3: Fig. 1.3 WiMAX Protocol Layers. Physical Layer protocols operates in three major frequency bands: to 66 GHz (licensed bands) 2. 2 to 11 GHz (licensed bands) 3. 2 to 11 GHz (license-exempt bands) To support these three bands, the protocol specifies multiple physical layers. Security Sub-layer or MAC Privacy Sub-layer This layer focuses on the security functions in the MAC layer. It consists of two component protocols: 1. Encapsulation Protocol: This component describes how the authentication is processed and the types of algorithms to be used in encrypting packets between the BS (Base Station) and the SS (Subscriber Station).

27 5 2. Key Management Protocol: This component describes how to distribute and manage connection keys. The default protocol used here is PKM (Privacy Key Management). Each connection between the BS and SS or MS (Mobile Station) has a unique CID (Connection ID). MAC common part sub-layer MAC layer is divided into two sub-layers. The common part sub-layer provides common functionality to all upper layer protocols. It is a connection oriented sub-layer and includes the mechanisms to request bandwidth. Authentication and registration is also a part of this layer's functionality MAC convergence sub-layer (service specific convergence sub-layer): The MAC convergence sub-layer implements different services on top of the common part sublayer. It is also responsible for bandwidth allocation and QoS [4]. 1.4 WiMAX Compared to 3G and WiFi WiMAX unlike 3G (third generation of wireless communication) has taken into consideration the flexibility of its deployment by defining a selectable channel bandwidth from 1.25MHz to 20MHz. WiMAX relies on Orthogonal Frequency Division Multiplexing (OFDM) and Orthogonal Frequency Division Multiple Access (OFDMA) as opposed to Code Division Multiple Access (CDMA) used in 3G. WiMAX can achieve higher spectral efficiencies than 3G especially because of the boosting accomplished by multiple antennas. OFDM makes it easier for WiMAX to exploit frequency diversity [5]. Typically in 3G systems the ratio between the upload and the download link is fixed, while WiMAX supports the flexibility of changing the downlink-to-uplink data rate ratio. For supporting various types of media traffic, WiMAX Media Access Control (MAC) is designed from the ground up to support different types of traffic such as real time,

28 6 constant bit rate, and variable bit rate traffic patterns. 3G also supports a variety of QoS levels. WiMAX has a clear advantage over 3G in terms of cost. Its simplified IP architecture puts it ahead on the price/performance curve compared to the complex 3G architectures. On the other hand, the WiMAX mobility support is somehow not yet proven to match the existing 3G systems. As WiMAX networks are starting to be deployed, their mobility will be proven in time. 1.5 WiMAX Standards IEEE standards group was formed in It focused initially on developing a standard for a point-to-multipoint Line-of-Sight (LoS) communication using 10 to 60 GHz. The first standard came out in December The group then added NLOS support in the new amendment called a using 2 to 11 GHz. Other features were introduced in that amendment like: the support of OFDM and several additions to MAC layer. A new standard, called also known as d, replaced all the old standards and presented the basis of the first WiMAX solution but it was for fixed subscriber stations. In December 2005, the group added the mobility support through their introduction of the new standard e. The e standard offers the support of mobile and nomadic subscriber applications. Table 1[5] below illustrates the basic data on IEEE Standards. Status e-2005 Completed Completed June Completed December December GHz for fixed and mobile applications Frequency Band 10GHz-66GHz 2GHz-11GHz 10GHz-66GHz Application Fixed LOS Fixed NLOS Fixed and mobile NLOS

29 7 MAC Architecture Transmission Scheme Modulation QPSK, 16QAM, 64QAM Gross Data Rate 32Mbps ~ 134.4Mbps Multiplexing Burst TDM/TDMA (Time Division Multiplexing) Duplexing TDD (Time Division Duplexing) and FDD (Frequency Division Duplexing) Channel 20MHz, Bandwidths 25MHz,28MHZ Air Interference Designation WiMAX implementation Point to Point to Multipoint, Multipoint, Mesh Mesh Single Carrier only Single Carrier, 256 OFDM or 2,048 OFDM WirelessMAN-SC None QPSK, 16QAM, 64QAM 1Mbps-75Mbps Burst TDM/TDMA OFDMA TDD and FDD 1.75MHZ, 3.5MHz, 7MHz, 14MHz, 1.25MHZ, 5MHZ, 10MHz, 15MHz, 8.75MHz WirelessMAN-SCa WirelessMAN- OFDM WirelessMAN- OFDMA WirelessHUMAN 256-OFDM as fixed WiMAX Point to Multipoint, Mesh Single Carrier, 256 OFDM or scalable OFDM with 128, 512, 1024, or 2048 subcarriers QPSK, 16QAM, 64QAM 1Mbps-75Mbps Burst TDM/TDMA OFDMA TDD and FDD 1.75MHZ, 3.5MHz, 7MHz, 14MHz, 1.25MHZ, 5MHZ, 10MHz, 15MHz, 8.75MHz WirelessMAN-SCa WirelessMAN- OFDM WirelessMAN- OFDMA WirelessHUMAN Scalable OFDM as Mobile WiMAX Table 1.1 WiMAX Standards WiMAX offers a convenient solution to the last mile problem. Its broadband speed capability, wireless nature and frequency diversity present it as a powerful competition to current cable and DSL connections. WiMAX is a work-in-progress project that is still being tested and updated continuously, which allows it to overcome the current

30 8 limitations of the existing 3G technologies. WiMAX has its advantages over 3G in many areas especially in performance/cost field. WiMAX is expected to compete with 3G technologies especially in 3 rd world country with a lot of rural areas such as China.

31 9 Chapter 2 WiMAX Physical Layer This chapter gives a brief introduction about the WiMAX Physical Layer and its various functionalities. In addition to that, it introduces the concepts and definitions used in this layer. 2.1 Introduction IEEE defines three PHY (Physical) layers for WiMAX: Single Carrier, Orthogonal Frequency Division Multiplexing (OFDM), and Orthogonal Frequency Division Multiple Access (OFDMA). WiMAX Single Carrier or WirelessMAN-SC PHY is intended to operate in the frequency range of 11GHz~66GHz. This standard assumes LOS (Line of Sight) operation due to the propagation characteristics [6]. Mobility in this standard is infeasible. WirelessMAN-OFDM defined by the d standard uses OFDM technology and is targeted to operate under NLOS conditions. Still, WirelessMAN-OFDM is for fixed communication only and does not support mobility. It operates on the frequency range below 11GHz. The last and most recent standard is WirelessMAN-OFDMA, which also operates on the frequency range below 11GHz, hence it does not require LOS operations. The main advantage of OFDMA is that it allows mobility. The multiple access support in WirelessMAN-OFDM is done using TDMA (Time Division Multiple Access) or DAMA (Demand Assigned Multiple Access). WirelessMAN-OFDMA uses OFDMA as its multi-access scheme [6][7].

32 Basic Concepts and Definitions This section aims to give the reader the necessary information about the terminology used in WiMAX world and the basic concepts behind WiMAX. It will cover mainly the latest standard for WiMAX, WirelessMAN-OFDMA or IEEE e, in more details. References are given if more clarifications are needed. FEC (Forward Error Correction) This technique is used to enhance the possibility of detecting errors on the other end of transmission and if possible to correct them. FEC adds extra bits to the transmitted date block (redundant data bits). The ratio of the actual data to the transmitted data is called code rate. FEC adds complexity to the transmitted system. The level of code rate used depends on the channel characteristics and status. Single Carrier and Multi-carrier When using a single carrier to transmit data at a high rate, ISI (Inter Symbol Interference) occurs due to delay spread, fading and the multipath characteristic. ISI effect can be reduced if longer symbol duration is used. To avoid sacrificing the desired high bandwidth multi-carrier modulation scheme is used. In this scheme, the available frequency is divided into subcarriers with much smaller signal rate. The subcarriers are usually orthogonal to each other to minimize cointerference. Duplexing, Multiplexing, and Multiple Access Duplexing indicates bidirectional communication between BS and SS. Duplexing can be categorized into full duplex, where transmission and reception operations are done in the same time, and half duplex, where only one operation is active. The other type of categorization is based on the domain the duplexing is done in. In TDD (Time Division Duplexing) the transmission (Downlink) and receiving (Uplink) duplexing is achieved by specifying a time slot for each operation on the same frequency. FDD

33 11 (Frequency Division Duplexing) assigns different frequencies to each operation to achieve duplexing. Fig 2.1 shows the difference between FDD and TDD. Fig. 2.1a TDD Fig. 2.1b FDD Multiplexing refers to the capability of a device to transmit to multiple devices. This can be achieved either using TDD by reserving a slot for each user, or using FDD by reserving a frequency for each. Multiple-Access indicates the ability of multiple devices to access the medium. There are main four types of multiple-access techniques. In Time Division Multiple Access (TDMA), the access to the medium is shared between the users by allocating different slots to each user depending on the allocating strategy. Frequency Division Multiple Access (FDMA) assigns different frequencies to each user. In OFDMA (Orthogonal FDMA), a subset of mutually orthogonal subcarriers are assigned to a device for a certain time to access the channel. CDMA or Code Division Multiple Access grants each user a different code to encode the transmissions. The design of CDMA targets minimizing autocorrelation and cross correlation between the used codes.

34 OFDM/OFDMA in WiMAX Since the architecture of the physical layer of WiMAX is based on OFDM technology, this section explains in details the benefits from using this technology and how it s integrated in the WiMAX architecture OFDM basics OFDM is based on the idea of dividing a high-bit-rate data stream into smaller low bitrate modulated streams called subcarriers. Theses subcarriers are mutually orthogonal which increases OFDM efficiency. Fig 2.2 shows the difference in frequency utilization between OFDM and FDM where OFDM is more efficient in utilizing channel bandwidth. Fig 2.2 Comparison between FDM and OFDM Efficiency OFDM has many advantages over other techniques, which promoted it to be the first choice for WiMAX standard. One of the main reasons behind favoring OFDM is its low computational complexity. OFDM can be easily implemented (both modulator and demodulator) using FFT (Fast Fourier Transform) / IFFT (Inverse FFT). It also eases the use of Adaptive Modulation and Coding (AMC), where the modulation of a channel is changed depending on the channel status. OFDM exploits frequency diversity, which can provide robustness again burst errors.

35 OFDMA Basics OFDM is multi-carrier modulation method but it does not support multi-access. OFDMA is a hybrid approach to allow multiple-access feature to the communication medium. OFDMA is a combination of TDMA and FDMA. Users are allocated time slots (Time Domain) and subcarriers (Frequency Domain). Subcarrier allocation takes into consideration the channel condition to allow best possible communication performance. OFDMA shares the OFDM frequency diversity and robustness. Because of its flexibility nature in both time and frequency domain, OFDMA can support various data rates and QoS constrains. Fig 2.3 OFDMA Frequency Allocation Figure 2.3 above shows how the base station can assign different frequencies per user, this allocation is done per time slot OFDM in WiMAX WiMAX uses OFDM in its two standards WirelessMAN-OFDM for fixed communication and WirelessMAN-OFDMA for mobile communication. Fixed

36 14 WiMAX uses FFT of size 256, while mobile WiMAX uses various FFT sizes from 128 till 2048 (except 256). Subcarriers are divided into three categories: data subcarriers, pilot subcarriers, and null subcarriers. Figure 2.4 shows the subcarriers distribution. Fig 2.4 Subcarrier Distribution In fixed WiMAX the FFT size is fixed, that implies that when the bandwidth changes the subcarrier spacing will vary. Increasing subcarrier spacing leads to decreasing the symbol time. In order to overcome the delay spread that leads to ISI (Inter Symbol Interference) more subcarriers are to be allocated as guard bands. Pilot signals or subcarriers are used as a reference to track channel quality. Guard band's sole purpose is to prevent interference and there is no power allocated to them. Table 2.1 shows the distribution of subcarriers in OFDM-256. Functionality Number assigned Data subcarriers 192 Pilot subcarriers 8 Guard band DC subcarrier 1 Total = 256 Table 2.1 Subcarrier Distribution for FFT-256 Mobile WiMAX uses scalable OFDMA, which supports a range of FFT sizes from 128 to Such flexibility helps overcoming the limitations of fixed WiMAX. When the bandwidth is increased; instead of increasing the guard band size, the FFT size can be increased to keep the subcarrier spacing fixed to khz.

37 OFDM Subchannelization Subchannelization is a technique used to group subcarriers into a unit called subchannel. This technique enhances the performance of the link. WiMAX-OFDM uses it in a limited fashion in the uplink. Where 16 subchannels are identified and the SS can be assigned 1, 2, 4, 8, or all the sets. WiMAX-OFDMA allows subchannelization in both ways (Uplink and Downlink). Subchannels can be formed using contiguous subcarriers, also called Adjacent Subcarrier Allocation (ASCA), or randomly distributed carriers, also called Distributed Subcarrier Allocation (DSCA). ASCA groups subcarriers with the highest SINR (Signal-to-Interference-Noise-Ratio), which is very helpful in many fixed application, but it lacks the support for mobile application. On the other hand, DSCA way of allocation maximizes frequency diversity. DSCA is considered the best approach for mobile environment.. Based on DSCA, there are several distribution schemes. The default WiMAX distribution profile is called Partially Usage of Subcarriers (PUSC). It defines 30 and 35 subchannels for downlink and uplink channels respectively for 10MHz channel bandwidth. AMC (Adaptive Modulation and Coding) is based on ASCA. Though it lacks the frequency diversity, it supports what is called multi-user diversity. This helps allocating subcarriers to users depending on their frequency response. 2.5 PHY Frame Structure The minimum time-frequency that can be allocated by WiMAX is called a slot. Slot is one subchannel in the frequency domain and one, two or three OFDM symbols in the time domain depending on the subchannelization scheme used. The contiguous slots

38 16 allocation for a user is called data region. Data regions are assigned based on demand, QoS parameters, and channel status. In TDD mode, OFDM and OFDMA frame (physical layer data unit) is divided into two subframes (downlink and uplink subframes). Figure 2.5 shows the structure of both downlink and uplink subframes. Notice that there is gap between the two subframes. This works as a guard band. This gap is called TTG (Transmit-Receive Transmission Gap), which separates the downlink and uplink subframes. Another Gap is called RTG (Receive-Transmit Transmission Gap) that separates the uplink and downlink subframes. WiMAX also supports FDD frames, where uplink and downlink frames are sent simultaneously. Fig. 2.5 OFDMA WiMAX Frame in TDD Mode

39 17 Most of the coming deployments of WiMAX will be likely using TDD because of its advantages over FDD. TDD supports more flexibility in sharing the bandwidth between uplink and downlink. And because it does not requires paired spectrum to support the two operations at the same time. TDD also has a simpler design than the one required for FDD. TDD suffers from the complexity of keeping the base stations synchronization to ensure there is no interference. 2.6 TDD Frame Components To ensure time and frequency synchronization a preamble precedes all WIMAX downlink frames. Preamble also contains the initial channel estimation. FCH (Frame Control Header) is used to provide the necessary control information like base station ID and the download burst profile (modulation and coding scheme) that the SS needs to know in order to decode the subframe [8]. Uplink and Downlink MAPs are used to specify the users' data regions in downlink and uplink subframes. MAPs also contain the burst profile for each subscriber. To ensure that the MAP messages are delivered, it is sent using a robust modulation such as BPSK (Binary Phase Shift Keying) with ½ coding rate. As shown in Fig 2.4, one downlink subframe can contain multiple bursts for different users. Although, IEEE allows frame sizes from 2ms to 20ms, initially all WiMAX equipments will support only 5ms frame. Uplink subframe is also made up of several bursts that belong to different users. A portion of the uplink subframe is reserved as a contention period, allowing new or mobile devices to request services. This portion is named Ranging Channel, and is used to determine the time offset to be used by different Subscriber Stations (SS).

40 WiMAX Physical Layer Blocks This section illustrates the role WiMAX physical layer plays to support the characteristics of WiMAX. The OFDMA and OFDM channel encoding as illustrated in Fig 2.6 consists of several steps before data transmission occurs. These steps are: randomization, channel coding, and interleaving. Fig 2.6 WiMAX Physical Layer Conceptual Diagram Randomization Randomization is a process to systematically or randomly reorder or randomize the transmitted data. It is employed to minimize the possibility of transmission of an un-modulated carrier and to ensure adequate numbers of bit transitions to support recovery. Randomization is achieved by XORing the data blocks with a pseudo-random binary sequence (PRBS) generated using a certain polynomial [6]. Another purpose of randomization is to encrypt the transmitted data blocks to prevent any unintended receiver from decoding the data.

41 19 Channel Coding As described earlier FEC (Forward Error Correction) process is used to maximize the possibility of detecting and possibly recovering the corrupted received data by adding redundancy to the transmitted data. WiMAX-OFDM standard specifies three methods of FEC: Reed-Solomon concatenated with convolutional coding (RS-CC), block turbo coding (BTC), and convolutional turbo coding (CTC). WiMAX-OFDMA specifies five methods of channel coding: convolutional coding (CC) with tail biting, block turbo coding (BTC), convolutional turbo coding (CTC), low density parity check coding (LDPCC), and CC with zero tailing. The most common channel coding method is CTC. Interleaving The encoded data from the pervious step go though a two-step process. The first step ensures that adjacent encoded bits are mapped into non adjacent subcarriers to provide frequency diversity and to improve the performance of the decoder. The second step maps the adjacent bits to the less and more significant bits of the constellation. The modulation of data bits depends on the modulation scheme used. WiMAX takes into consideration channel quality to choose the correct modulation scheme. The modulation scheme is selected per subscriber to achieve the best performance possible. The number of bits per symbol (time) depends on the modulation scheme used, for QPSK (Quadrature Phase Shifting Keying) it is 2, for 16- QAM (Quadrature Amplitude Modulation) it is 4, and for 64-QAM it is 6. Symbol Mapping The final sequence of data bits resulting from the previous steps is then converted to a sequence of complex data symbols. The complex data symbol vector takes the value I + jq. Figure 2.7 show QPSK modulation where each symbol is modulated using two bits.

42 20 Fig 2.7 QPSK Modulation Constellation Table 2.2 shows the values used by the symbol mapper to convert the streamed bits to its complex vector values. Here B(m) represents the modulation bit of a sequence to be transmitted, where m is the bit number (m ranges from 1 through n). For example, B(1) matches the first bit entering the modulator, and B(2) corresponds to the second bit entering the modulator. B(1) B(2) I Q Table 2.2 QPSK bits to Symbol Mapping Before modulation both I and Q signals are filtered using square-root raised cosine filters [13]. The ideal square-root equation is illustrated in Fig 2.8. H ( f ) = 1 for f fn(1 α) H ( f ) = H ( f ) = π sin 2 2 fn fn f α for fn(1 α) f fn(1 + α) for f Fn(1 + α) Fig 2.8 Square-root Filter Equation This filtering helps shaping the baseband pulse and reduces the harmonics generated at high frequencies.

43 Permutation Schemes In OFDMA PHY, the mapping from data bit to physical subcarriers is achieved in two steps. The first step is to map the data to one or more time slots and on one or more logical subchannels. The second step is called permutation, in which the logical subchannels are mapped to physical subcarriers. WiMAX-OFDMA defines seven permutation schemes: FUSC, PUSC, O-FUSC, O-PUSC, AMC, TUSC1, and TUSC2. Permutation zone is defined as the data region in both time and frequency domain that has the same permutation scheme. The two common permutation schemes in WiMAX DL frame are FUSC (Full Usage of Subcarriers) and PUSC (Partial Usage of Subcarriers). FUSC allocates the pilot subcarrier first in subchannels that are made up of 48 data subcarriers, then the reminder of the subcarriers are mapped onto subchannels. FUSC has two sets of pilots: fixed (pilot subcarriers index does not change) or variable (the index changes from one symbol to another). FUSC is designed to minimize the probability of using the same subcarriers in adjacent sectors or cells. PUSC is the default permutation in WiMAX. In PUSC, the subcarriers are grouped into clusters, each cluster is formed with 14 subcarriers and 2 OFDM symbols. That contains 24 data subcarrier and 4-pilot subcarrier. The clusters are then grouped to form six groups as shown in Table 2.3 for FFT size of 1,024. In each group the permutation process is done independently. A subchannel is created using two clusters from the same group. FFT Size 1024 Subcarriers per cluster 14 Number of subchannels 30 Data subcarriers used 720 Pilot subcarriers 120 Left-guard subcarriers 92 Right-guard subcarriers 91 Table 2.3 Parameters of PUSC permutation in OFDMA-DL

44 22 DL-PUSC is designed to minimize the probability of using the same subcarrier in adjacent sectors or cells. PUSC has an advantage over FUSC for its flexibility and better support for mobility. Figure 2.9 illustrates PUSC permutation process. Due to PUSC flexibility, all or just a subset of the 6 groups can be assigned to a specific transmitter. Segmentation in PUSC is the process of separating neighboring transmitters subcarriers which leads to tighter frequency reuse, but at the same time it affects the data rate. Fig 2.9 DL-PUSC Subcarrier Permutation Scheme AMC (Adaptive Modulation and Coding) is considered one of the implementation of adjacent subcarrier allocation ASCA. It provides a promising permutation alternative. Its main advantage is to rapidly assign the correct modulation and coding level (MCS) to each subscriber depending on the channel level. With AMC, subscribers that are close to the base station are typically assigned less robust modulation scheme with less redundancy in the transmitted data (e.g., 64 QAM with R=3/4 code ratio), but the

45 23 modulation-order and/or code rate will decrease as the distance from the base station increases. Fig 2.10 shows an example of such modulation distribution. Fig 2.10 WiMAX System Using AMC Permutation In this chapter we described the basic concepts and technologies used in WiMAX PHY layer. Starting with the description of the physical layer signal characteristics where the significance if using OFDM technology is showed. Then we explained the way WiMAX standard defines the steps and the data bits are processed in order to be transmitted. Finally, a short explanation of the most common permutation schemes used in WiMAX was presented.

46 24 Chapter 3 EESM This chapter describes the meaning of EESM and Beta (β) value. In addition to that it shows the role these parameters play in the determination of the condition of the communication channel. It also provides an explanation to the main purpose of this thesis. 3.1 Introduction In order to simulate WiMAX, we need to simulate 1024 (or more) subcarriers, the effect on noise on each of these subcarriers and their effect on the received FEC blocks. Such a simulation can be very complex and time consuming. This complexity can be avoided by modeling the channel as an additive white Gaussian noise (AWGN) channel with a single effective SINR. Wireless scientists have developed several ways to combine the SINRs of multiple subcarriers in to an effective SINR. One of the commonly used methods is the so called "Exponential Effective SINR Mapping" or EESM. This method is described below in detail. 3.2 EESM EESM (Exponential Effective SINR Mapping) is used to map the instantaneous values of SINRs to the corresponding BLER (Block Error Rate) value. Although EESM was introduced to work with SIR (Signal to Interference Ratio), it works with SNR as well. EESM is a simple mapping method used when all the subcarriers of a specific subscriber are modulated using the same Modulation and Coding Scheme (MCS) level. The basic idea of EESM is to find a compression function that maps the set of SINRs

47 25 to a single value that is a good predictor of the actual BLER [9]. Fig 3.1 shows the main purpose behind using EESM function. Here, BLER refers to block error rate and PER refers to packet error rate. Note that average SINR is not a good predictor of actual BLER or PER (Packet Error Rate).. SINR 1 SINR 2 SINR compression to effective SINR SINR eff Mapping to PER / BLER BLER/PER.. SINR N Fig 3.1 EESM Functionality EESM is a channel-dependent formula that maps power level as well as MCS level to SINR values in the AWGN (Additive White Gaussian Noise) channel domain. Such function allows its mapping along with AWGN assumptions (such as effect of increase in power, CINR/MCS threshold tables) to predict the effect of MCS and boosting modification. The method has been shown to yield an accurate estimation of the AWGN-equivalent SINR (henceforth referred to as effective SINR ) for frequency selective channels [11]. In case of multi-carrier transmission as in WiMAX, the set of subcarrier SINRs are mapped with the help of EESM formula into a scalar instantaneous effective SINR value. An estimate of the BLER value is then obtained, using the effective SINR value, from basic AWGN link-level performance. The mapping of the effective SINR value to the corresponding BLER value will use either a look-up table for the mapping function or use an approximate analytical expression if available. The EESM method estimates the effective SINR using the following formula (Figure 3.2):

48 26 γ eff 1 N EESM γ, β ) β ln e i = 1 N γ i ( β Fig 3.2 EESM Formula Where, γ is a vector [γ 1, γ 2,, γ N ] of the per-subcarrier SINR values, which are typically different in a frequency selective channel. β is the parameter to be determined for each Modulation Coding Scheme (MCS) level, and this value is used to adjust EESM function to compensate the difference between the actual BLER and the predicted BLER. To obtain β value, several realizations of the channel have to be conducted using a given channel model (e.g., Ped B and Veh A). Then BLER for each channel realization is determined using the simulation. Using the AWGN reference curves generated previously for each MCS level, BLER values of each MCS is mapped to an AWGN equivalent SINR. These AWGN SINRs for n realizations can be represented by an n- element vector SINR AWGN. Using a particular β value and the vector γ of subcarrier SINRs, an effective SINR is computed for each realization. For n realizations, we get a vector of computed effective SINRs denoted by Γ eff. The goal is to find the best possible β value that minimizes the difference between computed and actual effective SINRs: β = arg min SINRAWGN Γ eff β ( β ) Fig 3.3 Beta Calibration Formula

49 27 Previous works to achieve the corresponding β value for each MCS level has yielded close results. Table 3.1 shows the results obtained from [13], while Table 3.2 and Table 3.3 show the results obtained from [14]. Modulation Code Rate β 1/ QPSK 2/ / / QAM 2/ / Table 3.1 Beta Values from 3GPP Modulation Code Rate β QPSK 1/ / QAM 1/ / QAM 1/ / Table 3.2 Beta Values using Ped A Channel Model Modulation Code Rate β QPSK 1/ / QAM 1/ / QAM 1/ / Table 3.3 Beta Values using Ped B Channel Model There are three types of test environments to determine the channel quality: Indoor (like office environment), outdoor pedestrian and outdoor vehicular. And there are two main delay spread behaviors: low delay spread (A), and medium delay spread (B). The

50 28 following tables shows the delay and power values for (Ped A and Ped B) and (Veh A and Veh B) channel models provided by ITU-R Recommendation M Tap Channel A Channel B Relative Delay (ns) Average Power (db) Relative Delay (ns) Average Power (db) Table 3.4 Delay and Power value for Ped A and Ped B Channel Models Tap Channel A Channel B Relative Delay (ns) Average Power (db) Relative Delay (ns) Average Power (db) , , , , , , , Table 3.5 Delay and Power value for Veh A and Veh B Channel Models

51 29 Chapter 4 Simulation Setup and Results In this chapter, the simulation model used to obtain the results is explained in detail. The simulation parameters used are also presented. Each of the beta calibration steps is described in detail. Finally, the last section shows the results obtained from the simulation process. 4.1 Introduction Our physical layer simulation model is constructed using CoWare SPD (Signal Processing Designer), formally known as SPW. In order to get the desired β value for each MCS level, hundreds of channel realizations have to be conducted. Using CoWare SPD eases the simulation process and allows repeatable and consistent simulations outcomes. CoWare SPD is a C-based modeling and simulation environment that facilitates structured modeling and model reuse. CoWare SPD allows the users to create their own models or use any needed module from the provided libraries. Creating a model is done either by using C-based instructions or by utilizing MATLAB models. Such flexibility in the designing environment allows using many of the already proven code blocks. CoWare SPD includes a lot of useful libraries beside the basic signal process libraries. It provides several libraries to support the simulation of WLAN and WMAN systems. It includes IEEE system test benches for downlink and uplink communication models.

52 Downlink Model CoWare SPD downlink model consists mainly of four components as shown in fig 4.1. The four components are: IEEE downlink transmitter, IEEE channel, IEEE downlink receiver, and for computation purposes a BLER block Downlink Transmitter Fig.4.1 Downlink Simulation Model This block performs as a transmitter, where all the operations to the downlink frame data are performed. Then the FCH and the preamble sections are added. The bit stream is then modulated to produce the transmitted waveform. Fig 4.2 shows the main components of the downlink transmitter module. Fig.4.2 Downlink Transmitter Model

53 31 The five main components of the downlink transmitter model are: Randomizer, Preamble, Frame Control Header, Zone Transmitter Downlink and OFDM modulation. Each of these components ensures that the downlink frame is shaped corresponding to the user parameters supplied in the beginning of the simulation. Randomization This block is responsible to generate random bit stream to be considered as the net input data of the transmission process. Such block takes into consideration bit rate, sampling frequency and the probability of false (zero) bits in the stream. Preamble This block generates the downlink frame preamble; it depends on the IDCell, Preamble Index, and segment number. The data generated are converted to binary. Then it is modulated and mapped to OFDMA symbols. Frame Control Header FCH block produces the frame control header in the downlink frame. Since the information in this header is highly important, the downlink prefix is modulated using QPSK modulation with ½ coding rate and four repetitions. After that it is mapped into a burst. Zone Transmitter Downlink In this block the transmitted data are encoded and modulated. The modulated bits are mapped to the corresponding data carriers depending on the chosen permutation scheme. After that the pilot, guard and DC subcarriers are inserted. Randomization process is performed on all the subcarriers. OFDM Modulator This block performs IFFT transform of the subcarriers to produce the OFDMA time waveform. To insure the resistance against Inter Symbol Interference a circular extension is added to the resulted waveform.

54 32 The output waveform generated from the downlink transmitter is then passed to the channel model Channel Model CoWare SPD provides the support of the main channel models described in the IEEE standard with the flexibility to add other models. It supports the AWGN channel, fixed communication channel, and mobile ITU IMT-2000 pedestrian and vehicular channels. For ITU, it supports both delay spread A and B. The process is divided into two steps: first, the input signal is used to produce multiple versions of the signal at different delays. Then, the delayed signals are combined and weighted to produce the desired effect of the multi-path faded version of the input signal. Fig 4.3 shows the channel components. Fig.4.3 Channel Model Block The OFDM waveform passes another step where Additive White Gaussian Noise (AWGN) is then added to create the estimated Eb/Noand SNR ratios at the receiver Downlink Receiver and BLER Block To ease the simulation process, the downlink receiver is assumed to be ideal receiver. This means that the assumption of perfect synchronization with the received data is