WiMAX is the system for wireless broadband access. It is based on IEEE 802.16 standards which are mainly based on Orthogonal Frequency Division Multiplexing (OFDM) technology. OFDM is a wideband modulation scheme using multicarrier digital communication [Bahai & Saltzberg, 1999][Molisch, 2002]. During the communication there is an uplink and a downlink between base station transmitter and mobile receiver. While communicating, the wireless channel effects on the received OFDM signal differ due to selection of different parameter values as well as existing Signal to Noise Ratio (SNR) condition at that time. SNR variations may arise due to distance variations between transmitter and receiver and fast or slow mobility, exhibiting time varying channel [Lowray, 2001]. Different allowable bandwidths with same number of subcarriers, different number of subcarriers with same bandwidth, different modulation mapping schemes [Hole & Qien, 2001], variation in pilot power [Alsusa; Baidas & Lee, 2005] and pilot positions, pilot sequences, variations in cyclic prefix interval etc., are considered as important OFDM parameters requiring critical selection, whose effect is directly reflected in the performance. Up to certain extent adaptive nature is adopted in mobile WiMAX standard 802.16e in terms of scalable OFDMA [IEEE 802.16e, 2005] and linkage of modulation mapping scheme with channel coding [Hole and Qien, 2001] which can be extended further in terms of few more parameters. Such parameters are identified and reinvestigated. Few more possibilities are described here in a comprehensive manner. These parameters can be made adaptive with SNR conditions of the channel in both uplink as well as downlink. This chapter investigates those parameters which can be varied and on the basis of that adaptive OFDM transmission link control concept is developed, which can be applied in practice, maybe in vehicular mobility up to certain extent. Indirectly, effective and efficient Quality of Service (QoS) control can be achieved. Such control can be adopted in any frame based systems in general with at least half duplexity between transmitter and receiver. This adaptive nature may reduce the wastage of unnecessary energy utilized for the users who are very near to the transmitter, bringing the optimum solutions. Of course, the discussion reflects multiuser scenario.
After identifying such parameters we have simulated the mobile WiMAX for OFDM-256 case using different channel types (as this scheme is common to both fixed as well as mobile WiMAX) and BER-SNR plots for different such parameters, with given conditions, are represented in the chapter at the end to approximate the optimum SNR conditions for various such parameters to maintain the targeted BER.
Adaptive algorithms are also presented along with adaptive user allocation. These techniques utilize knowledge obtained by dynamically tracking the radio channel response, to optimize the user bandwidth and subcarrier modulation. Adaptive modulation independently optimizes the modulation scheme applied to each subcarrier so that the spectral efficiency is maximized, while maintaining a target BER [Hole and Quien, 2001]. The performance of this technique is dependent on the correlation of the frequency selective fading and how fast the fading changes with position of the transceiver.
2. WiMAX Scenario
The demand for high-speed mobile wireless communications and use of the radio spectrum is rapidly growing with terrestrial mobile communication systems being just one of many applications vying for suitable bandwidth. These applications require the system to operate reliably in non-line-of-sight environments with a propagation distance of 0.5 - 30 km, and at velocities up to 100 km/hr or higher. This operating environment limits the maximum RF frequency to 5 GHz, as operating above this frequency results in excessive channel path loss, and excessive Doppler spread at high velocity. This limits the spectrum available for mobile applications, making the value of the radio spectrum extremely high [Wu and Lin, 2006]. The Mobile WiMAX standards IEEE 802.16e onwards are developed by keeping in mind the above scenario. Obviously, mobile WiMAX defit in terms of bit rate compared to fixed WiMAX. To meet the demand of speed and spectrum the physical layer becomes very important. Indirectly, by physical layer optimization using parameter control, one can get the required Quality of Service (QoS). Physical layer is based on scalable OFDM in case of mobile WiMAX [IEEE 802.16e, 2005], where, OFDM technology promises to be a key technique for achieving the high data capacity and spectral efficiency requirements for wireless communication systems of even 4G. For the visualization of physical layer design and transmission control, one must know the architecture. WiMAX architecture is shown in Figure 1, which includes line of sight (LOS) and non line of sight (NLOS) communication links. Figure 2 shows the development of WiMAX system over cellular infrastructure, and adaptive modulation requirement in a cell.
Instead of fixed allocations of physical parameters over such architecture, adaptive nature in multiuser diversity has become a topic of recent interest and is mainly required in multiuser scenario. Adaptive user allocation exploits the difference in frequency selective fading between users, to optimize user subcarrier allocation. In a multipath environment the fading experienced on each subcarrier varies from user to user, thus by utilizing user/subcarrier combinations that suffer the least fading, the overall performance is maximized. Of course there must be the considerations for the minimum calculation complexity to meet the real time requirements.
3. OFDM vs Scalable OFDMA in WiMAX
OFDM and scalable OFDMA (SC-OFDMA) work slightly differently but both have the feasibility of adaptive nature. SC-OFDMA is akin to mobile version of WiMAX only, while former is in both the fixed as well as mobile WiMAX cases. In OFDM when we increase the bandwidth, we increase the channel bandwidth of each tone as the number of tones remain constant. But in SC-OFDMA we instead increase the number of Fast Fourier transform (FFT) points increasing the channel bandwidth, bandwidth of the tones is kept constant in the mobile environment. The FFT size and the number of carriers are equal in both fixed and mobile WiMAX, based on OFDM256, but they are different in SC-OFDMA. The fixed bandwidth is a compromised solution in mobile environment [though with such simpler case of 256 FFT point based OFDM simulations are done for time varying channels and results are given later on].
In SC-OFDMA, Scaling of FFT to the channel bandwidth is in order to keep the carrier spacing constant across different channel bandwidths (typically 1.25 MHz, 5 MHz, 10 MHz or 20 MHz). Constant carrier spacing results in higher spectrum efficiency in wide channels, and a cost reduction in narrow channels. Other bands not multiples of 1.25 MHz are defined in the standard, but because the allowed FFT subcarrier numbers are only 128, 512, 1024 and 2048, other frequency bands will not have exactly the same carrier spacing, which might not be optimal for implementations [IEEE 802.16e, 2005].
SC-OFDMA (used in 802.16e-2005) and OFDM256 (802.16d) are not compatible thus most equipment will have to be replaced if an operator wants or needs to move to the later standard. However, some manufacturers are planning to provide a migration path for older equipment to SC-OFDMA compatibility which would ease the transition for those networks which have already made the OFDM256 investment. Intel provides a dual-mode 802.16-2004 802.16-2005 chipset for subscriber units. This affects a relatively small number users and operators.With the advent of mobile WiMAX, there is an increasing focus on portable subscriber units. 802.16e-2005 has been accepted as IP-OFDMA for inclusion as sixth wireless link system under IMT-2000. So it is very important in the present scenario.
4. Adaptive Parameters Identification
Figure 3 shows ideal OFDM spectrum setting along with few subcarriers adjusted in a particular transmission bandwidth maintaining spacing f and orthogonality among the subcarriers. Frequency offset due to Doppler effect destroys the orthogonality [Bahai & Saltzberg, 1999]. Thus, OFDM based system has to satisfy four requirements in general while designing:
1) Available bandwidth: The bandwidth limit will play a significant role in the selection of number of subcarriers along with spacing. Large bandwidth will allow obtaining a large number of subcarriers with reasonable Cyclic Prefix (CP) length avoiding multipath.
2) Required bit rate: The size of the frame must be decided on the basis of symbol mapping scheme and number of subcarriers (hence spacing f) to be assigned to it. This will decide bit rate indirectly.
3) Tolerable delay spread based on terrain and distance: A user environment specific maximum tolerable delay spread should be known beforehand in determining the CP length.
4) Doppler spread based on velocity support: The effect of Doppler shift due to user movement should be taken into account for allowable subcarrier spacing.
The design parameters which can be applied adaptation are derived according to the system requirements. The transmission parameters are adjusted to provide an acceptable level of performance to the most impaired link. This approach then limits the performance that might be offered to subscribers with less impaired channels. Clearly this method results in sub-optimal utilization of the total channel capacity. Hence, the identified design parameters for an adaptive OFDM system are as follows with the opportunities for optimizing the overall system performance (For which the simulation is done):
1) Number of subcarriers and subcarrier spacing: It is stated earlier that the selection of large number of subcarriers (Of course within specified bandwidth) will help to combat multipath effects. But, at the same time, this will increase the synchronization complexity at the receiver side as well as increase problems due to Doppler spread. So, allocate user subcarriers so as to minimize Signal to Interference Ratio (SIR) in cellular systems and to minimize the effects of frequency selective fading.
2) Symbol duration and CP length: For specified delay spread perfect ratio between the CP length and symbol duration should be selected, so that multipath effects are combated and significant amount bandwidth is not lost due to CP.
3) Modulation type per subcarrier: The performance requirement will decide the selection of modulation scheme. Adaptive modulation can be used to support the performance requirements in changing environment. So, dynamically allocate the modulation scheme on an individual subcarrier basis to match the current channel conditions. In our simulations, by varying the SNR conditions the BER is tested for various modulation schemes to find out the limiting conditions. Of course, it is necessary to adopt appropriate channel coding in accordance with as given in Table 2 and 3.
4) Bandwidth: Dynamically change the bandwidth of each user based on the link quality. This allows the bandwidth of weak users to be reduced so that their energy spectral density remains sufficiently high to maintain communications. The concept of SNR-Bandwidth trade-off can be exploited.
There are many papers in which the adaptive allocations of pilot is proposed. Pilot management is necessary because it is an extra investment of power and that should be optimized. Here are few possibilities summarized.
Superimposed pilots over data symbols
Variation in Pilot positions (randomly hopping or sequentially varied), same way at the receiver.
Variation in the length of training sequence
Variation in the training sequence (pattern) itself. There is a separate research area in which sequences and their properties are analyzed and genetic algorithms are there to develop sequences with desirable properties.
Variation in the pilot power—maintaining the Peak to Average Power Ratio (PAPR)
Variation in the number of pilots- Again PAPR comes in picture
Apart from these, Forward Error Correction, adaptive power control and adaptive user allocation are important issues. Table 1 contains a synthetic view of some adaptive techniques used nowadays in broadband multicarrier wireless systems, including WiMAX, together with the benefits they bring.
5. Radio Resource Management by Adaptive Features
Advanced radio resource algorithms in broadband wireless systems enable service providers to maximize subscriber throughput and overall coverage while maintaining QoS. Techniques to optimize the use of available radio resources include power control, rate adaptation, automatic repeat requests, channel quality indication, scheduling, and admission control. WiMAX with its OFDMA-based structure provides a means to balance the effects of these techniques to provide an optimal tradeoff between throughput and link quality.
Adaptive power control is an important function for ensuring link quality. In the upstream direction, adaptive transmit power control is used to maximize the usable modulation level, which achieves the highest throughput, while at the same time controlling interference to adjacent cells. In the downstream direction, different power allocations for specific subchannels can be used to provide better service to subscribers at the edge of the cell while providing sufficient signal levels to subscribers in closer proximity to the base station.
Improved Power Consumption
The mobile WiMAX standard incorporates mechanisms that enable subscriber terminals to be active only at certain times as negotiated with the base station. When no data is to be transmitted or received, the subscriber terminal can move to ‘sleep’ or ‘idle’ modes to minimize power consumption. The base station scheduler is kept aware of every sleep or idle subscriber terminal and has the ability to switch the terminal to transmit or receive mode whenever required. In the subscriber terminal transmit mode the use of subchannels ensures that the transmit power is no greater than what is necessary to maintain sufficient link quality consistent with the traffic being transmitted, thus further reducing power consumption in the subscriber terminal.
In any terrestrial multi-cellular network, mobile subscribers will experience transmission path conditions that vary with relative location and time. With OFDMA the specific modulation and coding scheme can be adapted on a per subscriber basis dependent on path conditions to maximize channel throughput while maintaining link quality to each subscriber. With OFDMA systems, the subcarriers are modulated with either the more robust QPSK or the higher order, more efficient QAM modulations – with the more sophisticated modulation schemes having higher throughput but being much more susceptible to interference and noise. This rate adaptation, through adaptive modulation and error coding schemes ensures that the number of bits conveyed by each subcarrier is optimized relative to the CINR required to ensure a reliable air link connection. OFDMA systems can also increase throughput to individual subscribers by increasing the number of allocated subchannels at any given time. Both of these concepts are included in the mobile WiMAX specification.
Hybrid Automatic Repeat Request
Automatic repeat request (ARQ) algorithms are well known in wireless, and wireline, networks for retransmitting failed transmissions. The effective use of ARQ however, requires precise selection of both transmit power and data rate for the retransmissions, otherwise the link becomes underutilized or experiences excessive packet errors. Since it is challenging to maintain these optimal settings in the time varying environment of mobile broadband services, a significantly more robust mechanism called Hybrid- ARQ (H-ARQ) was developed. With H-ARQ, which is part of the mobile WiMAX specification, the receiver combines the information from a faulty packet with the re-transmissions of the same packet until enough information is gathered to retrieve the packet in its entirety.
Channel Quality Indication
Timely channel quality indication (CQI) messages at the receiver are essential for adaptive power and rate control and H-ARQ to be effective. The support of high mobility services requires that fast corrective actions be taken at the transmitter to ensure the link is operating optimally at all times. Mobile WiMAX specifies a compact size (4-6 bits each) CQI messages, resulting in lower delay and greater reliability than regular control messages. This ensures that the CQI messages provide fast and reliable feedback of path conditions to the base station while maintaining low overhead.
Scheduling control is a mechanism, located in the base station, for managing upstream and downstream packet allocations based on traffic requirements and channel conditions at any given moment. The scheduler allocates radio resources in frequency and time, based on considerations such as; QoS parameters for the specific traffic-type, individual subscriber service level agreements (SLA), and connection-by-connection path conditions. Since data-oriented traffic can vary considerably between uplink and downlink, asymmetric capacity allocation is also supported in time division duplex (TDD) implementations with appropriate radio resources and packet assignments done on a per-sector basis for a variable duration based on actual demand. These basic scheduling control mechanisms are part of the mobile WiMAX standard.
Admission control is the process of determining whether or not to allow a new connection to be established based on: current traffic conditions, available resources, and cumulative QoS requirements. Excessive traffic in a cell increases the amount of interference to adjacent cells thus reducing cell coverage. Admission control is used to accept or reject the connection requests so as to maintain the cell load within acceptable limits. The admission control function is located in either the WiMAX base station or the access service network (ASN) gateway where the load information for several base stations can be monitored.
6. Trade off Between Allocated Bandwidth and Adaptive Bandwidth
In most cases user is allocated a fixed amount of bandwidth, regardless of the received signal power. But in mobile WiMAX like environment or in multiuser scenario, this may lead to problems for users that have low received signal strength [Lowray, 2001]. The SNR of these users may be insufficient to support communications even using BPSK. The SNR seen at the receiver is dependent on the signal bandwidth, and so reducing the bandwidth while using the same transmitter power increases the SNR of the signal. For example, reducing the signal bandwidth by 5 times, allows the full transmitter power to be concentrated into one fifth the bandwidth, increasing the transmitted power spectral density by 5 fold, resulting in an improved received SNR of 5 dB.
The main aim of adaptive bandwidth allocation is to maintain communications with users that have low received signal strength due to far distance with respect to the base station transmitter. This is achieved by reducing their bandwidth to the point where the transmitted power spectral density is high enough to support communications at a low data rate. This can be used as a method for improving the quality of service (i.e. decreasing the outage probability).
Adaptive bandwidth by itself will not be suitable for all applications, especially those that required a fixed data rate such as streaming video and audio. In these applications a joint optimization of bandwidth and modulation scheme could be performed to maintain a fixed data rate, while minimizing the amount of bandwidth used at any one time. This could be achieved by allocating both the user bandwidth and modulation scheme so that the spectral efficiency multiplied by the user bandwidth results in the required data rate. This way, as the signal strength becomes weaker, the amount of bandwidth allocated to that user increases to compensate. This fixed data rate optimization is not included in the simulations and could be researched.
7. Trade off between Fixed Modulation and Adaptive Modulation
Adaptive modulation has not been used extensively in wireless applications due to the difficulty in tracking the radio channel effectively. Work has been done studying the use of adaptive modulation in single carrier systems, however not many works have been published on use of adaptive modulation in OFDM systems.
Most OFDM systems use a fixed modulation scheme over all subcarriers for simplicity. However each subcarrier in a multiuser OFDM system can potentially have a different modulation scheme depending on the channel conditions. Any coherent or differential, phase or amplitude modulation scheme can be used including BPSK, QPSK, 8-PSK, 16-QAM, 64-QAM, etc, each providing a trade off between spectral efficiency and the bit error rate. The spectral efficiency can be maximized by choosing the highest modulation scheme that will give an acceptable BER.
In a multipath radio channel, frequency selective fading can result in large variations in the received power of each subcarrier. For a channel with no direct signal path this variation can be as much as 30 dB in the received power resulting in a similar variation in the SNR. In addition to this, interference from neighboring cells can cause the SNR to vary significantly over the system bandwidth.
To cope with this large variation in SNR over the system subcarriers, it is possible to adaptively allocate the subcarrier modulation scheme, so that the spectral efficiency is maximized while maintaining an acceptable BER. Figure 4 shows a reference plot for applying adaptive modulation to an individual subcarrier as the channel SNR varies with time.
Using adaptive modulation has a number of key advantages over using static modulation. In systems that use a fixed modulation scheme the subcarrier modulation must be designed to provide an acceptable BER under the worst channel conditions. This results in most systems using BPSK or QPSK. However these modulation schemes give a poor spectral efficiency (1 - 2 b/s/Hz) and result in an excess link margin most of the time. Using adaptive modulation, the remote stations can use a much higher modulation scheme when the radio channel is good. Thus as a remote station approaches the base station the modulation can be increased from 1 b/s/Hz (BPSK) up to 4 - 8 b/s/Hz (16-QAM – 256-QAM), significantly increasing the spectral efficiency of the overall system. Using adaptive modulation can effectively control the BER of the transmission, as subcarriers that have a poor SNR can be allocated a low modulation scheme such as BPSK, or none at all, rather than causing large amounts of errors with a fixed modulation scheme. This significantly reduces the need for Forward Error Correction
There are several limitations with adaptive modulation. Overhead information needs to be transferred, as both the transmitter and receiver must know what modulation is currently being used. Also as the mobility of the remote station is increased, the adaptive modulation process requires regular updates, further increasing the overhead.
There is a trade off between power control and adaptive modulation. If a remote station has a good channel path the transmitted power can be maintained and a high modulation scheme used (i.e. 64-QAM), or the power can be reduced and the modulation scheme reduced accordingly (i.e. QPSK). Distortion, frequency error and the maximum allowable power variation between users limit the maximum modulation scheme that can be used. The received power for neighboring subcarriers must have no more than 20 - 30 dB variation at the base station, as large variations can result in strong signals swamping weaker subcarriers. Inter-modulation distortion results from any non-linear components in the transmission, and causes a higher noise floor in the transmission band, limiting the maximum SNR to typically 30 - 40 dB. In our simulations the SNR is limited to 25 dB. Frequency errors in the transmission due to synchronization errors and Doppler shift result in a loss of orthogonality between the subcarriers. A frequency offset of only 1 - 2 % of the subcarrier spacing results in the effective SNR being limited to 20 dB.
7.1 Adaptive Modulation and Coding Support in Mobile WiMAX
Schemes offering varied modulation and coding for different classes of channel propagation conditions by assigning user classes to different physical channels are a step in the direction of improved channel utilization, but do not completely address the dynamic nature of network traffic and its impact on required link gains. The PHY layer described as per adaptive modulation and coding optimizes channel utilization by permitting dynamic adaptation of both modulation format and Forward Error Correction (FEC) rate on a subscriber-by-subscriber basis. Thus, if the channel characteristics vary over time, say seasonal variations in path loss caused by the presence or absence of foliage, that subscriber’s modulation and coding parameters can be adjusted to compensate for these changes.
WiMAX supports a variety of modulation and coding schemes and allows for the scheme to change on a burst-by-burst basis per link, depending on channel conditions. Using the channel quality feedback indicator, the mobile can provide the base station with feedback on the downlink channel quality. For the uplink, the base station can estimate the channel quality, based on the received signal quality. In conditions of good signal, a highly efficient 64QAM coding scheme is used, whereas when signal is poorer, a more robust BPSK scheme is used. QPSK and 16QAM are for intermediate conditions. Following is a list of the various modulation and coding schemes supported by WiMAX-802.16e as well.
The key parameter for traffic modeling (maybe using adaptive antenna) from a planning tool point of view is the bit rate that can be offered for a given received signal level. This characteristic itself depends on the modulation used, hardware algorithms, transmitting and receiving antenna specifications.
PHY-Layer Data Rates:
Because the physical layer of WiMAX is quite flexible, data rate performance varies based on the operating parameters. Parameters that have a significant impact on the physical-layer data rate are channel bandwidth and the modulation and coding scheme used. Other parameters, such as number of subchannels, OFDM guard time, and oversampling rate also have an impact. Operating over maximum range (50 km) increase bit error rate and thus must use a lower bit rate. Lowering the range allows a device to operate at higher bit rates. Following is the PHY-layer data rate at various channel bandwidths, as well as modulation and coding schemes with scalable OFDM cases.
Further, the OFDM symbol is one block of frame expanded to frame duration during serial to parallel conversion and assigned to a subcarrier (usually a block is referred to as an “OFDM symbol”) A ‘block adaptive’ PHY is proposed for 802.16.3 in which modulation format and coding rate can be adjusted on a block by block basis to provide optimum channel utilization for the widest range of channel conditions. This proposal is offered as a PHY layer for the 802.16.3 Task Group [IEEE 802.16.3c-00/39, Oct 2000] The ‘Block-Adaptive’ PHY is based upon a Frequency Division Duplex (FDD) approach employing Downstream-Upstream access methods of DOCSIS like TDM data distribution in the downstream path and Time Division Multiple Access (TDMA) in the upstream direction. Although it is envisioned that many traffic models will require asymmetric data rates in the downstream and upstream directions, the PHY layer proposed here allows for symmetric data rates if so desired. Further, the proposed PHY can support a wide range of channel bandwidths and assignment plans. In the “Block-Adaptive” structure the data is partitioned into Blocks delimited by a Block Identifier Word. Each Block represents a unit of data to be sent to an individual subscriber.
7.2 Adaptive Modulation Based TCP Aware Uplink Scheduling in IEEE 802.16 Netwrks
There are many schemes at higher layers utilizing the adaptive allocations of PHY. One of the scheme is highlighted here as a case study. The primary contribution of paper by Hemant Kumar Rath, Abhay Kharandikar and Vishal Sharma is to propose a fair adaptive modulation-based uplink scheduling scheme for applications based on TCP in IEEE 802.16. Since, the TCP congestion window size (cwnd) changes only after one RTT, cwnd is an indication of the number of time slots required per Round Trip Time (RTT). Hence, instead of assigning equal number of slots to all users, we argue that the BS should assign slots in proportion to their cwnd, i.e., as per the flow’s requirement. Assigning time slots based only on cwnd will result in unfairness among the TCP flows, since flows with smaller RTT s will have larger window size as compared to the flows with larger RTT. To avoid this unfairness, they introduced a credit-based approach that ensures fairness among the flows. More slots are assigned to the flows which are closer to their TCP timeout, thereby preventing their congestion window from dropping to one due to timeout. By introducing adaptive modulation, fairness measure that only considers slots assigned becomes irrelevant, rather, fairness in terms of amount of data transmitted in a frame should be considered. Hence, they measured fairness on the amount of data transmitted by SSs.
8. Tracking Schemes for Experimentation [Lowrey, 2001]
a. Adaptive Subcarrier Allocation Algorithms:
There are several methods for allocating subcarriers to users in a multiuser OFDM system on adaptive basis. The main five schemes are to use a group of subcarriers with
A fixed frequency grouped subcarriers
Randomly hopped subcarriers
Spread out subcarriers in a comb pattern
Adaptive user allocation
These methods are described here to have experimentation. For performing such experiment, set up with multiple OFDM transceivers is necessary for measurements of path loss as a function of frequency and distance. This data can be used for the simulations. Perfect channel estimation is necessary to assume in this simulation. All five simulations can use adaptive modulation in order to determine the modulation for each subcarrier in each time slot, with the only difference being the subcarrier frequencies allocated to each user. The adaptive modulation and the subcarrier frequency allocation can be updated at a rate of 250 Hz, which corresponds to a distance traveled between updates of 4% of wavelength, for the carrier frequency of 1000 MHz used. The frequency allocations can be subdivided into 256 or less subcarriers over the system bandwidth of 1.25-3.5 MHz.
In practice the number of subcarriers would be more than this, however the number should be kept reasonably low so that the frequency allocations could be seen in the simulation plots. A BER threshold of 1x10-5 can be used to decide on the SNR thresholds for the adaptive modulation allocation. The choice of this BER threshold is arbitrary and dependent on requirements of the final application of the system. Changing this threshold would result in a slight change in data throughput,
In each of these simulations multiple users are transmitting at the same time in the same frequency band. These multiple transmissions form a single OFDM signal at the base station’s receiver. In order for these signals to remain orthogonal to each other, they must all be frequency and time synchronized with each other. The time synchronization must be accurate to within the effective guard period length (shorter then the actual guard period length due to channel delay spread), while frequency synchronization must be sufficiently accurate as to maintain a sufficiently high effective SNR as to use the modulation schemes used in the simulation.
b. Adaptive Bandwidth Algorithm:
1) Initially allocate all users an equal number of subcarriers, using the adaptive user allocation algorithm (described next)
2) Calculate the SNR of all the subcarriers for a given user, based on the bandwidth allocated and the particular subcarriers that were allocated in step 1.
3) Find the minimum SNR of the subcarriers allocated to each user.
4) For each user check that every subcarrier allocated to that user, has a SNR greater than a given threshold. This is done to ensure that each subcarrier has a sufficiently high SNR to support at least BPSK modulation. A threshold of 5-6 dB can be used in the simulations from the observations. If a user has any subcarriers below this threshold then reduce the BW (number of subcarriers) allocated to that user. This frees up BW, making it available to other users. Make sure that each user is allocated at least one subcarrier. This prevents the degenerate case where a user is allocated zero subcarriers, which in turn makes the estimated SNR infinite (The SNR is inversely proportional to the signal bandwidth).
5) Redistribute the free BW by allocating it to users that have a SNR significantly greater than the minimum SNR threshold. Transmitting over a wider BW results in a lower transmitted spectral density and received SNR. Thus there is no use reallocating the extra BW to a user that is just above the SNR threshold, as this would result in their SNR dropping below the threshold, defeating the purpose of the reallocation. To help with data load leveling, the extra BW should be allocated to users that have a low SNR first, as users with a high SNR will already have a comparatively high data rate and thus shouldn’t need more BW.
6) Based on the new user BW allocations, reallocate the subcarriers to users by repeating the adaptive user allocation algorithm
7) Repeat from (2) until all allocated subcarriers have a SNR above the required threshold.
If all subcarriers meet the minimum SNR threshold, then the user allocation is complete, so apply adaptive modulation to each subcarrier and exit. If the SNR of one or more subcarriers is below the threshold then continue onto step 5.
c. Adaptive User Allocation Algorithm:
1) Find the mean SNR over the entire system BW for each user.
2) Perform allocation of subcarriers to users in-order, from, lowest mean SNR, to highest mean SNR. This helps to ensure that weak users get access to the best subcarriers.
3) Sort the SNR response for the user being allocated, removing any subcarriers that have already been allocated to other users. The SNR response is the SNR of each subcarrier as seen by that user. This will be different for each user, due to propagation variations.
4) Allocate from the sorted SNR response subcarriers in descending order from best SNR to worst SNR subcarriers to meet the BW required for the user. Repeat from step 2 until all users have been allocated.
9. Simulation and Results
The OFDM modulation scheme based WiMAX system simulation is done using MATLAB and the following results are obtained for finding the limiting conditions to eliminate outages. From the plots the concept of feasibility of adaptive allocations is clear.
Same guard ratio ¼ but different bandwidth (Figure 5) – here due to cyclic prefix addition the IFFT points increases and hence the spectrum widens slightly. More the bandwidth more effect of noise, so the probability of errors increases and hence with wider bandwidth, the performance degrades. At the same time Doppler margin will also come in picture. When the bandwidth is increased with the same number of carriers and FFT points, effectively the Doppler margin between carriers increases. This is rather advantageous in heavy Doppler condition. After addition of CP the original spectral setting does not get affected. With the above discussion it can be summarized that the bandwidth related performance is a tradeoff. It can be identified from the graphs that bandwidth 1.25 give the best performance in all mapping schemes. That is why the majority of the previous plots are taken with 1.25 MHz bandwidth with and ¼ guard ratio. Figure 6. gives the different SNR requirements for various types of channels. SNR requirement increase in this order: AWGN, SUI 1, ITU pedestrian A, SUI2-3, SUI 4, MIMO with Doppler, ITU pedestrian B, SUI 5-6 and ITU vehicular A. From the results of Figure 7, it can be commented that reducing the guard interval spectrum width reduces so effect of noise can reduce however, if the CP duration is less than CIR than complete mulitpath effect elimination is not possible. So depending upon the situation the performance improves or degrades. Consistency may not be there due to time varying nature of the channel, estimation and interpolation errors, random nature of the data etc. However, it is sure that if channel conditions are not known, the performance with guard ratio ¼ is optimum in most of the cases.