Peak-to-average power ratio management for multi-carrier modulation in wireless communication systems

ES 2 397 839 T3

DESCRIPTION

Management of the ratio between maximum and average powers for multi-carrier modulation in wireless communication systems

BACKGROUND

I. Field

The present invention relates generally to data communication and more specifically to techniques for managing the maximum to mean power ratio (PAPR) for multi-carrier modulation in wireless communication systems.

II. Background

Wireless communication systems are widely deployed to provide various types of communication, such as voice, data, etc. These systems can be multiple access systems capable of supporting the communication of multiple users, sharing the available resources of the system (for example, bandwidth and transmission power). Examples of such multiple access systems include code division multiple access (CDMA) systems, time division multiple access (TDMA) systems, frequency division multiple access (FDMA) systems, and frequency division multiple access systems. Orthogonal Frequency Division (OFDMA).

A wireless communication system can use multi-carrier modulation for data transmission. Common examples of multi-carrier modulation include orthogonal frequency division multiplexing (OFDM) and discrete multi-tone (DMT). OFDM effectively divides the overall bandwidth of the system into a number of orthogonal subbands. Each subband is associated with a respective carrier on which the data can be modulated. The carriers for the subbands can be independently modulated with data, and the modulated carriers are then grouped together to generate an output wave.

Multi-carrier modulation has certain desirable characteristics, including the ability to combat multi-path effects. However, a major drawback of multi-carrier modulation is that the maximum to mean power ratio (PAPR) for the output wave, that is, the ratio between the maximum power and the mean power of the wave generated by the modulation multi-carrier, can be high. High PAPR is a result of possible in-phase (or coherent) clustering of all carriers when independently modulated with data. In fact, it can be shown that the maximum power can be up to N times higher than the average power for multicarrier modulation, where N is the number of carriers.

The high PAPR for the waveform generated by multi-carrier modulation normally requires that the power amplifier be operated at a medium power level, which is usually much lower than the maximum power level (i.e., lowered from maximum power). . This is because large peaks in the waveform can cause the power amplifier to operate in a highly non-linear, or possibly clipped region, which would then cause intermodulation distortion and other phenomena that can degrade signal quality. By operating the power amplifier at a lower power than the maximum, where the recoil is usually between 4 and 7 dB, the power amplifier can address large maximum values ​​in the wave without generating excessive distortion. However, recoil represents ineffective power amplifier operation during other times, when large peaks are not present in the waveform. Therefore, it is highly desirable to minimize the PAPR of the waveform so that the power amplifier can be operated closer to the maximum power level, if desired or necessary.

Various schemes have been introduced to minimize PAPR for multi-carrier modulation. Most of these schemes attempt to reduce the PAPR of the same wave. For example, a conventional scheme proposes to correlate the data to be transmitted with specific code words that have been specially selected because they are associated with low values ​​of the PAPR. Another conventional scheme proposes to use "peak reduction carriers" that are modulated in such a way as to reduce the peak values ​​in the wave. Yet another conventional scheme proposes modulating the data on all carriers, but with different phases, to try to reduce the PAPR of the wave. These various conventional schemes for reducing PAPR may not be applicable for certain multi-carrier communication systems. This may be the case, for example, if data for all carriers is not available or accessible, as described below.

The document “Digital Video Broadcasting (DVB); Interaction Channel for Digital Terrestrial Television (RCT) incorporating Multiple Access OFDM; ETSI EN 301 958 ”, ETSI STANDARDS, LIS, SOPHIA ANTIPOLIS CEDEX, FRANCE, vol. BC, n ° V1.1.1, March 1, 2002 (2002-03-01), XP014004074 ISSN: 0000-0001 ”; refers to backward channel RF (radio frequency) link offsets, and service ranges.

Document EP 0 869 647 A2 refers to a multi-carrier modulation system with operating parameters

ES 2 397 839 T3 dynamically scalable.

Finally, document WO 98/15153 refers to a method and apparatus for mitigating the effects of intermodulation in multiple signal transmission systems. In one embodiment, the communication signals are allocated so that the high power signals are located near a center of the frequency band, while the low power signals are located near the edges of frequency bands.

There is, therefore, a need in the art for techniques to manage PAPR for multi-carrier modulation in wireless communication systems.

ABSTRACT

The invention is defined in independent claims 1 and 14. Techniques for managing PAPR in various multi-carrier and multiple access wireless communication systems (eg OFDMA systems) are provided herein. It is recognized that different terminals in a multiple access communication system may be associated with different required transmission powers in order to achieve their desired received signal qualities. Carriers can be assigned to terminals based on their required transmit powers.

In one aspect, the number of carriers to be allocated to each terminal is made dependent on its required transmission power. Fewer carriers can be allocated to a terminal with a higher required transmit power. Since a smaller PAPR is associated with a wave generated with fewer carriers, the power amplifier can be operated with a smaller recoil and the wave can be transmitted at a higher power level. On the contrary, more carriers can be allocated to a terminal with a lower required transmit power. Even though a higher PAPR is associated with a waveform generated with more carriers, the power amplifier can provide more recoil, since the required transmit power is lower.

In another aspect, the specific carriers to assign to the terminals are determined by their transmit power levels. Terminals with higher required transmitting powers are more likely to generate high levels of intermodulation distortion. These terminals may have carriers assigned near the middle of the operating band, so that their distortion may fall within the operating band. In contrast, terminals with lower required transmit powers are likely to generate low levels of intermodulation distortion. These terminals may have carriers assigned near the edges of the operating band, as the distortion will likely be below the specified out-of-band emission requirements.

Various aspects and embodiments of the invention will be described in greater detail below.

BRIEF DESCRIPTION OF THE DRAWINGS

The characteristics, nature and advantages of the present invention will become more apparent from the detailed description set forth below, when considered in conjunction with the drawings, in which identical reference characters identify correspondingly in their entirety, and in which:

Figure 1 shows a diagram of a multiple access wireless communication system;

Figure 2 shows a subband / carrier structure that can be used for an OFDMA system;

Figure 3A shows two hypothetical waves generated by multi-carrier modulation;

Figure 3B shows the transmission of the two waves with a maximum transmission power P_max and in a way that minimizes intermodulation distortion;

Figure 3C shows the transmission of the two waves, given the maximum transmission power Pmax and using the power control to achieve the desired quality of the received signal;

Figure 4 shows the assignment of carriers to terminals in such a way as to reduce out-of-band emissions;

Figure 5 shows a process for allocating and assigning carriers to terminals; Y

Figure 6 shows a block diagram of an access point and two terminals.

DETAILED DESCRIPTION

Figure 1 shows a diagram of a multiple access wireless communication system 100 employing multi-carrier modulation. System 100 includes a number of access points 110 that communicate with a number of terminals 120 (only two access points 110a and 110b are shown in Figure 1, for simplicity). An access point is a fixed station used to communicate with terminals. A point of

ES 2 397 839 T3 access may also be called a base station, or some other terminology.

A terminal is a station that communicates with the access point. A terminal can also be called an access terminal, user terminal, remote station, mobile station, wireless communication device, or some other terminology. Each terminal can communicate with one, or multiple, access points over the downlink and / or uplink, at any given time. Downlink (ie forward link) refers to transmission from access point to terminal, and uplink (ie reverse link) refers to transmission from terminal to access point.

A system controller 130 is coupled to the access points and may additionally couple to other systems, or other networks (eg, a packet data network). System controller 130 provides coordination and control to access points attached to it. Through the access points, the system controller 130 further controls the routing of data between the terminals, and between the terminals and other users coupled with the other systems or networks.

The techniques described herein for managing PAPR can be implemented in various multi-access and multi-carrier wireless communication systems. For example, system 100 may be an OFDMA system that uses OFDM for data transmission. Also, these techniques can be used for the uplink as well as the downlink. For clarity, these techniques are specifically described for the uplink in an OFDMA system. In the following description, an active terminal is one that is programmed for data transmission on the uplink (and possibly the downlink).

Figure 2 shows a subband / carrier structure 200 that can be used for an OFDMA system. The system has an overall system bandwidth of W MHz, which is divided into N orthogonal subbands 210 using OFDM. Each sub-band has a bandwidth of W / N MHz, and is associated with a respective carrier 212, on which the data can be modulated.

In a typical OFDM system, only M of the N total carriers are used for data transmission, where M <N. The remaining N - M carriers are not used for data transmission and their associated subbands serve as guard subbands for allow the system to satisfy spectral mask requirements. The M usable carriers include the carriers F to F + M - 1, where F is an integer usually selected so that the M usable carriers are centered in the middle of the operating band.

For OFDM, up to N carriers for the N subbands can be independently modulated with data. The modulated carriers are then grouped together to form an output wave. Modulated carriers can be grouped congruently (that is, in phase), in which case there will be a large amplitude in the wave. It can be shown that the maximum power of the wave generated with N independently modulated carriers can be many times greater than the mean power of the wave. The exact value for the PAPR depends on many factors. Also, the value of interest is often not the absolute maximum value, but some statistical value, for example, what value of instantaneous power is exceeded, say, 99% of the time.

Figure 3A shows graphs of two hypothetical waves 310 and 312 that are generated by multi-carrier modulation. The horizontal axis indicates time and the vertical axis indicates power. Wave 310 is generated with L carriers, and wave 312 is generated with 2-L carriers, where L can be any integer greater than one. The mean power of wave 310 is the same as that of wave 312. However, the maximum power of wave 312 is twice that of wave 310, because twice as many carriers were used to generate wave 312. Consequently, the PAPR of wave 312 is greater than the PAPR of wave 310.

A wave generated by multi-carrier modulation is usually transmitted in such a way as to limit the amount of intermodulation distortion. This requires the power amplifier for the wave to operate at a medium power level, P_avg, which is lowered or lowered from the maximum power level, P_max, for the power amplifier. The amount to be lowered is selected so that the power amplifier does not operate (or operate minimally) in a highly non-linear region, or clipping. More specifically, the kickback is normally selected so that the distortion generated by the power amplifier is limited to a specific level.

Figure 3B shows the transmission of the two waves in Figure 3A with the maximum transmission power P_max, and in such a way as to minimize distortion. Wave 310 can be transmitted with a BO reverse₁, which is determined, in part, by the PAPR₁ for this wave (e.g. BO₁ <PAPR₁). Similarly, wave 312 can be transmitted with a BO retrace₂, which is determined, in part, by the PAPR₂ for this wave (e.g. BO₂ <_PAPR₂). The mean transmission power (P_avg1) of wave 310 can be approximately twice the mean transmit power (P_avg2) of wave 312, while still limiting the distortion to approximately the same level. The exact ratio between P_avg1 And p_avg2 it depends on the specific retracements used for waves 310 and 312.

For an OFDMA system, the M usable carriers can be shared between multiple active terminals. In the link

ES 2 397 839 T3 upstream, each active terminal can be assigned a specific set of carriers on which it can transmit data. Both the number of carriers to be assigned to each active terminal, and which specific carriers to assign to the terminal, can be determined as described below. The carriers assigned to each terminal may or may not be contiguous. Each active terminal can then transmit using its specific assigned carriers.

Referring again to Figure 1, the terminals can be scattered throughout the entire length of the system. Each terminal is associated with a specific path loss to its access point, which is highly dependent on the distance between the terminal and the access point. Each terminal also requires a specific quality of signal received at the access point to achieve a desired level of performance. The required quality of the received signal can be quantized by a specific received signal-to-noise ratio (SNR), and the desired level of performance can be quantized by a specific frame error rate (FER), a packet error rate (PER), etc. The transmission power required for each terminal depends on its path loss and its required received signal quality.

If the terminals are scattered throughout the system, then the path loss is usually different from one terminal to another. Furthermore, the desired quality of the received signal may differ from one terminal to another depending on, for example, their data rates. Thus, the required transmission power is usually different from one terminal to another. In general, terminals that are located further from the access point have higher path losses to the access point and would then require higher transmission powers to achieve a given quality of received signal. For example, terminals 120a, 120b, 120d, and 120g will likely require more transmit power than terminals 120c, 120e, and 120f to achieve the same received signal quality at their respective access points.

Each terminal is associated with a specific maximum transmit power, P_max, which can be used for data transmission. This maximum transmit power can be determined by regulatory restrictions, system design, and / or limitations of the power amplifier used by the terminal. The maximum amount of transmit power that can be used for uplink data transmission would then be limited to Pmax.

A power control loop can be maintained to control the transmit power of each active terminal. Because there can be a large disparity in path loss for active terminals, the powers received at the access point for these terminals can vary greatly (for example, by as much as 80 dB) if these terminals are all transmitting at the same time. same power level. Even though orthogonal subbands are generated by OFDM, uplink transmissions from active terminals can interfere with each other due, for example, to timing and / or frequency offsets. To limit the amount of interference to nearby carriers, the transmit power of each active terminal can be controlled or adjusted so that the received signal quality for the terminal is within an acceptable range. The transmit power required for each terminal would then be determined based on the uplink power control, which can be large.

In one aspect, the number of carriers to be allocated to each active terminal depends on its required transmit power. Thus, different numbers of carriers can be assigned to different terminals, according to their required transmitting powers. Higher transmit power is required to achieve the desired received signal quality when the path loss is higher. If more transmit power is required, then fewer carriers can be allocated. Since a smaller PAPR is associated with a wave generated with fewer carriers, the power amplifier can be operated with a smaller recoil and the wave can be transmitted at a higher power level. On the contrary, since less transmit power is required when the path loss is smaller, more carriers can be allocated. Even though a higher PAPR is associated with a wave generated with more carriers, the power amplifier can provide more backlash, since the transmit power required for the wave is lower.

Figure 3C shows the transmission of the two waves in Figure 3A, given the maximum transmission power P_max, and using power control to achieve the desired received signal quality. Wave 310 is transmitted with a required average power P_req1, which is lowered by at least BO₁ from P_max. Wave 312 is transmitted with a required average power P_req2, which is lowered by at least BO₂ from P_max. The required average powers P_req1 And p_req2 They can be determined by the path losses and the required received signal qualities associated with the terminals transmitting these waves. The higher average power required for wave 310 may be due to higher path loss and / or higher received signal quality required for the wave. Kickbacks BO₁ and BO₂ they can be determined based on the PAPRs of these waves, as described above.

As shown in Figure 3C, for a power amplifier restricted by maximum transmit power P_max, the highest average power required P_req1 for wave 310 it can be provided by the power amplifier, since this wave is generated with fewer carriers and is associated with a smaller recoil. Even though wave 312 is generated with more carriers and is associated with greater recoil, the required mean power P_req2 for this wave it can also be provided by the power amplifier, since this power level is lower.

ES 2 397 839 T3

The maximum number of carriers that can be allocated to each active terminal can thus be made dependent on the required transmit power and the maximum transmit power for the terminal. The determination of the maximum number of carriers that can be assigned to each terminal can be done based on various schemes, two of which are described below.

In a first carrier allocation scheme, a table is formed for the maximum allowed average power with respect to the number of carriers. This table can include an entry for every possible number of carriers that can be assigned. For example, the table may include N entries for N carriers, where i indicates the number of carriers for the ith entry in the table. For each input, the highest average power P is determined_mavgi which can be used for the associated number of carriers, i (eg, empirically, by simulation, or by some other means). This maximum allowed average power, P_mavg¡i, is based on a hypothesis of the maximum transmit power P_max for terminals (which can be specified for the system or by regulatory restrictions). The table can be formed as shown in TABLE 1.

TABLE 1

<td>Number of carriers</td><td>Maximum allowed average power</td>

<td>N</td><td>^Pmavg, N</td>

<td>N</td><td>N</td>

<td>i</td><td>P mavg, i</td>

<td>N</td><td>= N</td>

<td> 1</td><td>^Pmavg, 1</td>

Since the waves with more carriers are associated with greater retracements, the maximum allowed average power decreases with increasing numbers of carriers (that is, P_mavg1 &gt; P_mavg2 &gt;&gt; P_mavg,_N).

The maximum number of carriers that can be allocated to each active terminal can then be determined based on the required transmit power, P_req, for the terminal and the table. In particular, the required transmit power for the terminal can be compared with the maximum allowed average powers in the table. The smallest maximum allowable mean power (P_mavg¡s) that is greater than or equal to P_req, and the number of carriers S associated with this P is determined_mavgS. The terminal can then be assigned any number of carriers less than or equal to

S.

In a second carrier allocation scheme, a table is formed for required setbacks with respect to the number of carriers. This table can also include an entry for each possible number of carriers that can be assigned. For each input, the retracement BO is determined_i minimum required for the associated number of carriers, i (eg, empirically, by simulation, or by some other means). This table can be formed as shown in TABLE 2.

TABLE 2

<td>Number of carriers</td><td>Backspace required</td>

<td>N</td><td>^bon</td>

<td>N</td><td>N</td>

<td>i</td><td>BO,</td>

<td>N</td><td>N</td>

<td> 1</td><td>BO₁</td>

The waves with the most carriers are associated with the largest retracements, so that BO_N &gt; ... BO₂ &gt; BO_i.

The maximum number of carriers that can be allocated to each active terminal can then be determined based on the required transmit power and the maximum transmit power for the terminal. In particular, the difference between the maximum and required transmit powers for the terminal is calculated first. This calculated difference is then compared to the required setbacks in the table. The largest recoil required (BOS) that is less than or equal to

ES 2 397 839 T3 that this calculated difference is then identified, and the number of carriers S associated with this BO is determined_S. The terminal can then be assigned any number of carriers less than or equal to S.

The maximum number of carriers that can be allocated to each active terminal can be initially determined, as described above. The effective number of carriers to be allocated to each terminal can then be determined based on any number of additional factors. Such factors may refer to (1) the amount of data to be transmitted, (2) fairness, (3) the priority of the terminals, and so on. The specific number of carriers actually allocated to each terminal is equal to or less than the maximum number of carriers that can be allocated. The specific carriers allocated to each terminal may or may not be contiguous.

Many wireless communication systems are operated in frequency bands with spectral mask requirements that limit the amount of out-of-band emissions. For these systems, the carriers to be assigned to each active terminal can be selected so that out-of-band emissions are reduced or minimized as much as possible.

Figure 4 shows typical emission requirements for a typical radio frequency (RF) operating band. The operating band has spectral mask requirements that are characterized by in-band specific maximum emission and out-of-band specific maximum emission. The maximum emission in the band can be specified, for example, by a specific transmission power restriction per MHz. Similarly, the maximum out-of-band emission can be specified by a specific transmission power restriction per MHz, below frequency f and above frequency f.₂.

Power amplifiers are typically designed so that they are linear at low to medium output power levels and become more non-linear at higher output power levels. Thus, when a power amplifier is operated at a high output power level, a higher level of non-linearity in the power amplifier can cause intermodulation distortion that falls outside the signal band. The amount of distortion depends on the specific design of the power amplifier and the output power level. If the non-linearity and / or the output power level are high enough, then the resulting distortion may exceed the specified maximum out-of-band emission requirement.

In another aspect, the specific carriers to be assigned to the active terminals are determined by their required transmit powers. A terminal with a higher path loss to the access point (for example, a terminal located near the edge of the coverage area) and / or a higher required received signal quality needs to transmit at a higher power level to achieve the quality of received signal required at the access point. This terminal is therefore more likely to generate a high level of intermodulation distortion. The terminal can then be assigned carriers near the middle of the operating band, so that the distortion can fall within the operating band. The higher level of distortion from this terminal can cause additional interference to other carriers, so the transmit powers for these carriers can be increased accordingly to compensate for the higher level of interference.

In contrast, a terminal with a smaller path loss to the access point (for example, a terminal located close to the access point) and / or a lower required received signal quality can transmit at a lower power level and achieve still the required quality of signal received at the access point. This terminal is therefore likely to generate a low level of intermodulation distortion. The terminal can then be assigned carriers near the edges of the operating band, since the distortion will likely be below the specified requirement for maximum out-of-band emission. The specific carriers allocated to each terminal may be located within a specific part of the operating band, but they do not need to be contiguous.

Figure 4 also shows the assignment of carriers to active terminals, in such a way as to reduce out-of-band emissions. A group of 410 carriers near the middle of the operating band can be assigned to a terminal that needs to transmit at a high output power level. Two groups of carriers 412 and 414 near the edges of the operating band can be assigned to the same, or different, terminals that can transmit at a low output power level. The uplink transmissions on these groups of carriers are from multiple terminals. However, these uplink transmissions are overlaid on the same graph in Figure 4, for clarity.

In a carrier assignment scheme, carriers are assigned to active terminals based on their required transmit powers. For a given transmission interval, the number of carriers to be allocated to each active terminal is first determined (eg, based on the required transmission power of the active terminal and possibly other factors, as described above). Active terminals can be associated with different required transmitting powers. The group of carriers to be assigned to the active terminal with the highest required transmit power is then selected to be near the middle of the operating band, the group of carriers for the active terminal with the next highest required transmit power is selected so that are closest to the middle of the operating band, and so on, and the group of carriers for the active terminal with the power of

The lowest required transmission is then selected to be near the edges of the operating band. This carrier allocation scheme can reduce out-of-band emissions as much as possible.

In another carrier assignment scheme, each usable carrier is associated with a respective power threshold level, and the carriers are assigned to the active terminals based on the power threshold levels and the transmitter powers required for the terminals; in particular, a given carrier can be assigned to a terminal if the required transmit power is equal to or less than the power threshold level. Carriers near the middle of the operating band can be associated with the highest power threshold levels, and those near the edges of the band can be associated with the lowest power threshold levels. These power threshold levels can be selected so that the specified out-of-band emissions can be satisfied for a given multi-carrier modulation scheme. Thus, a terminal located near the edge of the coverage area and with a higher required transmission power can only be assigned carriers near the middle of the operating band, while a terminal with a lower required transmission power can be assigned carriers can be assigned anywhere within the operating band.

Carriers can also be assigned to active terminals in some other ways to reduce out-of-band emissions, and this is within the scope of the invention. Furthermore, the carrier allocation techniques described herein can be used alone or in combination with the carrier allocation techniques described above.

Figure 5 shows a flow chart of one embodiment of a process 500 for allocating and allocating carriers to active terminals. Initially, the pertinent information regarding the transmission power of each terminal to be scheduled for data transmission is obtained (step 512). In one embodiment, the required and maximum transmit powers are obtained for each terminal. The transmission power required for each terminal can be sent by the terminal or obtained on the basis of some other means. The maximum transmission power for each terminal can be sent by the terminal, known a priori, or obtained on the basis of some other means. In another embodiment, the difference between the maximum and required transmit powers for each terminal is obtained. In yet another embodiment, the maximum transmit power and initial transmit power can be obtained for each terminal (for example, during registration), and the required transmit power for the terminal can then be estimated based on the transmit power initial and an accumulation of all power control commands sent to the terminal. Relevant transmit power information can therefore be provided in various ways.

The maximum number of carriers that can be allocated to each terminal is then determined based on the transmit power information (eg, based on the required and maximum transmit powers) (step 514). This can be accomplished using various schemes, such as the two carrier allocation schemes described above. A specific number of carriers are then allocated to each terminal based on (1) the maximum number of carriers that can be allocated to the terminal, (2) the total number of carriers available for allocation to all terminals, and (3) any number from other factors (step 516). The number of carriers awarded to each terminal is limited by the maximum number that can be awarded. Furthermore, the sum of all the carriers awarded to the terminals is limited by the total number of carriers available for the award.

Specific carriers are then assigned to each terminal, such that the amount of out-of-band emissions can be reduced or minimized (step 518). This can be accomplished using various schemes, such as the two carrier assignment schemes described above. The carriers assigned for each terminal can then be signaled to the terminal by a carrier assignment. Each planned terminal would then transmit using the specific assigned carriers and for the planned period of time.

Figure 6 shows a block diagram of one embodiment of an access point 110x and two terminals 120x and 120y in a multi-carrier, multi-access communication system 100.

On the downlink, at access point 110x, a transmission data processor (TX) 614 receives traffic data (i.e., information bits) from a data source 612, and signaling and other information from a controller 620 and a 630 planner. For example, the controller 620 can provide power control (PC) commands that are used to adjust the transmit power of the active terminals, and the scheduler 630 can provide carrier assignments for the terminals. These various types of data can be sent over different transport channels. TX data processor 614 encodes and modulates received data using multi-carrier modulation (eg, OFDM) to provide modulated data (eg, OFDM symbols). A transmitter unit (TMTR) 616 then processes the modulated data to generate a downlink modulated signal which is then transmitted from an antenna 618.

At each of the terminals 120x and 120y, the transmitted downlink modulated signal is received by an antenna 652 and provided to a receiver unit (RCVR) 654. The receiver unit 654 processes and digitizes the received signal to provide samples. A received data processor (RX) 656 then demodulates and decodes the samples to provide decoded data, which may include retrieved traffic data, messages,

ES 2 397 839 T3 signaling, etc. Traffic data can be provided to a data sink 658, and PC and carrier assignment commands sent to the terminal are provided to a controller 660.

Controller 660 directs data transmission on the uplink, using the specific carriers that have been assigned to the terminal, and indicated in the received carrier assignment. Controller 660 further adjusts the transmit power used for uplink transmissions based on received PC commands.

For the uplink, at each active terminal 120, a TX data processor 674 receives traffic data from a data source 672, and signaling and other information from controller 660. For example, controller 660 may provide information indicative of the required transmit power, the maximum transmit power, or the difference between the maximum and required transmit powers for the terminal. The various types of data are encoded and modulated by the TX data processor 674 using the assigned carriers, and further processed by a transmitter unit 676 to generate an uplink modulated signal which is then transmitted from the antenna 652.

At access point 110x, the uplink modulated signals transmitted from the terminals are received by antenna 618, processed by receiver unit 632, and demodulated and decoded by RX data processor 643. Receiver unit 632 can estimate received signal quality (eg, received signal-to-noise ratio (SNR)) for each terminal and provide this information to controller 620. Controller 620 can then obtain the PC commands for each terminal so that the received signal quality for the terminal remains within an acceptable range. The RX data processor 634 provides the recovered feedback information (eg, required transmit power) for each terminal to controller 620 and scheduler 630.

The scheduler 630 uses the feedback information to perform a number of functions such as (1) selecting a set of terminals for uplink data transmission and (2) assigning carriers to the selected terminals. The carrier assignments for the planned terminals are then transmitted on the downlink for these terminals.

For clarity, techniques for managing PAPR have been described specifically for the uplink in an OFDMA system. These techniques can also be used for downlink transmission, from the access point to the terminals. In a downlink transmission scheme, OFDMA is used for the downlink, similar to that for the uplink, and carrier multiplexing can be used to transmit data to multiple terminals on the downlink simultaneously within a given time slot. . In another downlink transmission scheme, data is transmitted to one terminal at a time, in a time division multiplexed (TDM) fashion. For both downlink transmission schemes, the number of carriers to assign to each terminal and the specific carriers to assign to each terminal can be determined as described above, based on the required transmit power for the terminal. For the OFDMA downlink transmission scheme, the available carriers can be assigned to multiple terminals, so that the downlink signal PAPR for all planned terminals is kept within a specific desired value. For the TDM-OFDM downlink transmission scheme, the number of carriers assigned to the served terminal can be selected so that the PAPR of the downlink signal to this terminal is also kept within the desired value. The data for each scheduled terminal can then be transmitted using the specific assigned carriers and with the required transmit power for the terminal.

For OFDM, the data to be transmitted by each carrier is first modulated (ie, symbol-correlated) using a specific modulation scheme selected for use for that carrier, in order to provide a modulation symbol for each symbol period. The modulation symbols for each terminal are then scaled to achieve the required transmit power for the terminal. Unused carriers are provided with zero signal values. For each symbol period, M scaled symbols, for M usable carriers, and N - M zeros for unused carriers, are transformed to the time domain using an inverse fast Fourier transform (IFFT), to obtain a symbol "transformed "Which includes N samples of the time domain. To combat inter-symbol interference caused by frequency selective fading (which is the result of a multi-path channel), a part of each transformed symbol can be repeated to form a corresponding OFDM symbol. The OFDM symbols generated in this way for different symbol periods are then processed to generate the downlink modulated signal that is transmitted to the terminals.

For the downlink, if the M usable carriers are all transmitted at the same power level, then the PAPR of the OFDM wave may be large. However, by assigning more carriers to terminals with lower required transmitting powers and fewer carriers to terminals with higher required transmitting powers, the PAPR of the wave will be smaller. This would then allow the power amplifier at the access point to be operated with less recoil and with a higher level of output power. This, in turn, may allow higher data rates to be used for one or more of the terminals.

ES 2 397 839 T3

The techniques described herein for managing PAPR for multi-carrier modulation can be implemented by various means. For example, these techniques can be implemented in hardware, software, or a combination thereof. For a hardware implementation, the elements used to implement the techniques at the access point and at the terminal can be implemented within one or more application-specific integrated circuits (ASIC), digital signal processors (DSP), devices digital signal processing (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), processors, controllers, microcontrollers, microprocessors, other electronic units designed to perform the functions described herein, or a combination thereof.

For a software implementation, the techniques described herein can be implemented with modules 10 (eg, procedures, functions, etc.) that perform the functions described herein. The software codes can be stored in a memory unit (for example, memory units 622 and 662 in Figure 6) and be executed by a processor (for example, controllers 620 and 660 and scheduler 630). The memory unit may be implemented within the processor or external to the processor, in which case it may be communicatively coupled with the processor by various means, as is known in the art.

The above description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be immediately apparent to those skilled in the art, and the generic principles defined herein can be applied to other embodiments without departing from the scope of the invention.