HK1148141A - Multicarrier communication system employing explicit frequency hopping - Google Patents

Description

Multi-carrier communication system using explicit frequency hopping

Technical Field

The present invention relates generally to multicarrier communication systems, and more particularly to Orthogonal Frequency Division Multiplexing (OFDM) systems employing frequency hopping.

Background

Frequency hopping is a spread spectrum technique used in many radio communication applications. In a frequency hopping spread spectrum system, a transmitter changes its transmission frequency over time according to a pseudo-random frequency hopping pattern. In effect, a transmitter "hops" from one frequency to another during transmission to spread its signal over a wide frequency band, while at any given moment the transmitted signal occupies a narrow frequency band. A hop period, referred to herein as a time slot, is a time interval during which the frequency is held constant. The frequency hopping pattern comprises a sequence of frequencies over which the transmitter hops.

Frequency hopping provides frequency diversity, which helps mitigate the effects of multipath fading if the spacing between subcarriers is large enough that fading is uncorrelated at different frequencies. Most mobile communication systems apply channel coding at the transmitter side and corresponding channel decoding at the receiver side. In order to exploit the frequency diversity provided by frequency hopping, the coded information block should be spread out over multiple hops (i.e., multiple time slots).

Frequency hopping may be used to share radio resources among multiple users. In conventional frequency hopping systems, different mobile terminals within the same cell or sector of a mobile communication system are assigned mutually orthogonal frequency hopping patterns so that the mobile devices do not transmit simultaneously on the same frequency in the same time slot. One way to ensure that the hopping patterns are orthogonal to each other is to use the same basic hopping pattern for all mobile devices, with different frequency offsets for each mobile terminal.

Different non-orthogonal frequency hopping patterns are typically used between cells, which means that simultaneous transmissions from two mobile devices in adjacent cells in the same frequency band during the same time slot may occur. When this happens, "collisions" occur, which means high interference levels during the corresponding time slots. However, since the channel coding spans several hops, the channel decoder is still typically able to decode the information correctly.

Frequency hopping may be applied in an Orthogonal Frequency Division Multiplexing (OFDM) system. In an OFDM system, one wideband carrier is divided into a plurality of subcarriers. A fast fourier transform is applied to the modulation symbols to spread the modulation symbols over a plurality of subcarriers of the wideband carrier. Frequency hopping can be implemented in an OFDM system by changing the subcarrier allocation.

Recently, there has been interest in using variable bandwidth allocations in the uplink of OFDM systems. The basic idea is to vary the bandwidth allocated to a mobile terminal based on its instantaneous channel conditions, buffer level, quality of service (QoS) requirements, and other factors. A scheduler in the network schedules the mobile terminals and determines their bandwidth allocations.

Frequency hopping has not previously been used in OFDM systems employing variable bandwidth allocations. One difficulty in applying frequency hopping techniques to OFDM systems that allow variable bandwidth allocations is that the number of available hopping patterns varies according to the bandwidth allocation. Furthermore, when mixing transmissions from two or more mobile devices using different bandwidths within one subframe (FDMA), the hopping probability of each mobile device depends on the bandwidth allocated to the other mobile devices. Another problem is that bandwidth allocation is dependent on the instantaneous channel conditions of the mobile device and therefore cannot be known in advance. If the frequency pattern is established without regard to bandwidth allocation, bandwidth allocation must be made to avoid collisions, which reduces the efficiency of the system.

Therefore, new scheduling techniques are needed to implement frequency hopping in OFDM systems that allow for variable bandwidth allocation.

Disclosure of Invention

The present invention provides a method and apparatus for implementing frequency hopping in an OFDM system that allows variable bandwidth allocation to mobile terminals. Variable bandwidth allocation is achieved by dynamically allocating different numbers of sub-carriers to different mobile terminals according to their instantaneous channel conditions. The frequency hopping pattern is determined "on-the-fly" based on the current bandwidth allocation for the simultaneously scheduled mobile terminals. The bandwidth allocation and frequency hopping pattern are signaled to the mobile terminal in a scheduling grant. Because the frequency hopping pattern is not predefined, the scheduling grant explicitly (explicit) signals the bandwidth allocation and frequency offset for each time slot within the scheduling interval.

The present invention provides a very flexible, simple (low complexity), and low overhead method to implement uplink frequency hopping in systems that support flexible bandwidth transmission.

Drawings

Fig. 1 shows an exemplary transmitter for implementing single carrier OFDM with variable bandwidth and frequency hopping.

Fig. 2 illustrates an exemplary OFDM processor for a single carrier OFDM transmitter.

Fig. 3 shows the structure of an exemplary OFDM carrier.

Fig. 4 illustrates an exemplary frequency hopping pattern for a single mobile terminal.

Fig. 5 shows mutually orthogonal frequency hopping patterns for two mobile terminals.

Fig. 6 illustrates how variable bandwidth allocation affects the available hopping patterns.

Fig. 7 shows an exemplary frequency hopping pattern in combination with variable bandwidth allocation.

Fig. 8 illustrates an exemplary access node in a mobile communications network that includes a scheduler for determining bandwidth allocations and frequency hopping patterns.

Fig. 9 illustrates an exemplary method implemented by a scheduler for scheduling uplink transmissions in a mobile communication system.

Detailed Description

Referring now to the drawings, an exemplary transmitter in accordance with an exemplary embodiment of the present invention is shown and indicated generally by the numeral 10. The transmitter 10 is configured to implement a transmission scheme known as single carrier orthogonal frequency division multiplexing (SC-OFDM). Variable bandwidth allocation and frequency hopping are used to achieve efficient utilization of radio resources. Variable bandwidth allocation is achieved by dynamically allocating different numbers of sub-carriers to different mobile terminals according to their instantaneous channel conditions. The frequency hopping pattern is determined "on the fly" based on the current bandwidth allocation. The bandwidth allocation and frequency hopping pattern are signaled to the mobile terminal in a scheduling grant.

Referring to fig. 1, a transmitter 10 includes a transmit signal processor 12, an Orthogonal Frequency Division Multiplexing (OFDM) processor 14, and a transmitter front end 16. The transmit signal processor 12 generates coded and modulated signals for transmission to the remote terminal. The transmit signal processor 12 may use any known form of modulation, such as Quadrature Amplitude Modulation (QAM) or Quadrature Phase Shift Keying (QPSK). An OFDM processor 14 receives the modulated signal from the transmit signal processor 12 and applies OFDM modulation to generate a transmit signal. The functions of the transmit signal processor 12 and the OFDM processor 14 may be implemented by one or more digital signal processors. The transmitter front end 16 is coupled to a transmit antenna 18. The transmitter front end 16 comprises a digital-to-analog converter for converting the transmission signal into analog form and radio frequency circuitry for filtering and amplifying the transmission signal.

Fig. 2 shows an exemplary OFDM processor 14 implementing a form of OFDM transmission known as single carrier OFDM (SC-OFDM). The components shown in fig. 2 represent functional elements that may be implemented by one or more processors. The OFDM processor 14 includes a Discrete Fourier Transform (DFT) module 22, subcarrier mapping circuitry 24, an Inverse Discrete Fourier Transform (IDFT) module 26, and a Cyclic Prefix (CP) module 28. A block of M modulation symbols of any modulation alphabet is input to a DFT module 22 of size M. The DFT module 22 performs DFT on the modulation symbols to convert the modulation symbols from the time domain to the frequency domain. The mapping circuit 24 maps the frequency samples output by the DFT module 22 to corresponding inputs of an IDFT module 26 of size N, where N > M. The unused input of IDFT module 26 is set to zero. IDFT module 26 transforms the frequency samples back to the time domain. In some embodiments of the invention, bandwidth expansion and spectral shaping (not shown) may be applied to the frequency samples in the frequency domain before conversion back to the time domain. For example, the spectral shaping circuit may be applied by multiplying the frequency domain samples with a spectral shaping function (e.g., a root-raised cosine function). The transmitted signal corresponding to a single block of modulation symbols is referred to herein as an OFDM symbol. Cyclic prefix module 28 then applies a cyclic prefix to the OFDM symbol.

Single carrier OFDM shown in fig. 2 may be considered OFDM with DFT-based precoding, where each IDFT input corresponds to one OFDM subcarrier. Thus, the term "DFT spread OFDM" or "DFTs-OFDM" is commonly used to describe the transmitter structure of fig. 2. The use of DFT-based precoding gives the last transmitted signal a "single carrier" characteristic, which means that each modulation symbol is "spread" over the entire transmission bandwidth and the transmitted signal has a relatively low peak-to-average power ratio compared to normal OFDM transmission. Assume a sample rate f at the output of IDFT module 26_sThen the nominal bandwidth of the transmitted signal will be BW-M/N · f_s。

The OFDM transmitter 10 shown in fig. 1 allows variation of an instantaneous transmission bandwidth by changing a block size M of modulation symbols input to the DFT module 22. Increasing the block size M will increase the instantaneous bandwidth required for transmission, while decreasing the block size M will decrease the instantaneous bandwidth required for transmission. Further, by shifting the IDFT input to which the DFT output is mapped, the transmitted signal can be shifted in the frequency domain.

Fig. 3 shows a structure of an exemplary OFDM carrier for uplink transmission. The vertical axis in fig. 3 represents the frequency domain, and the horizontal axis represents the time domain. In the frequency domain, the radio resources are divided into a plurality of narrowband subcarriers. A typical OFDM carrier may include hundreds or even thousands of subcarriers. In the time domain, the radio resources are divided into time slots. Each slot includes a plurality of symbol periods. In this example, one slot includes seven (7) symbol periods. One symbol period in each slot is used to transmit pilot symbols. The remaining six symbols in each slot are used to transmit data and/or control signals. The subcarriers in a slot may be grouped into units called resource blocks. For example, in the exemplary embodiments disclosed herein, one resource block includes twelve (12) subcarriers over a period equal to one slot.

For uplink scheduling, uplink radio resources are divided in the time domain into scheduling units called subframes. One subframe includes two or more slots. In the exemplary embodiment described herein, one subframe includes two (2) slots, but a different number of slots may be used. During each subframe, an access node (e.g., a base station) in a mobile communication network may schedule one or more mobile terminals for transmission on the uplink. The access node indicates the scheduled mobile terminal by sending a scheduling grant on a downlink control channel.

In some systems, variable bandwidth allocation in combination with orthogonal multiplexing schemes may be used to improve system throughput. In an OFDM system, it may not be efficient to allocate the entire available bandwidth to a single mobile terminal during a given time slot. The data rate that can be achieved by a mobile device is likely to be limited by the available power of the mobile device. Allocating the entire available bandwidth to power-limited mobile devices would result in a waste of system resources. When the mobile device cannot use the entire available bandwidth, a smaller transmission bandwidth may be allocated to the mobile device and the remaining bandwidth may be allocated to another mobile terminal. Thus, orthogonal multiplexing schemes such as Frequency Division Multiplexing (FDM) may be used to share the available bandwidth between two or more mobile terminals.

According to the invention, frequency hopping may be used in combination with variable bandwidth allocation in order to improve the robustness of the transmitted signal to fading, thus reducing bit errors that may occur during transmission. In a frequency hopping system, a transmitter changes its transmission frequency over time, for example, according to a pseudo-random frequency hopping pattern. Fig. 4 shows a frequency hopping pattern over twelve resource blocks and twelve time slots. As shown in fig. 4, a transmitter "hops" from one frequency to another during transmission to spread its signal over a wide frequency band, while at any given moment the transmitted signal occupies a narrow frequency band. In an OFDM system, frequency hopping may be implemented by shifting the frequency location of resource blocks allocated to mobile terminals during a scheduling interval. For example, if the scheduling interval used is one subframe, then different resource blocks in each slot within the subframe may be allocated to the mobile terminal.

In conventional frequency hopping systems, different mobile terminals within the same cell or sector of a mobile communication system are assigned mutually orthogonal frequency hopping patterns so that the mobile devices will not transmit simultaneously on the same frequency in the same time slot. One way to ensure that the hopping patterns are orthogonal to each other is to use the same basic hopping pattern for all mobile devices, with different frequency offsets for each mobile terminal. Fig. 5 illustrates how frequency hopping can be used to share available bandwidth between two or more mobile devices. As shown in fig. 5, each mobile terminal uses the same frequency hopping pattern. However, mobile device 2 has an offset of 3 resource blocks relative to mobile terminal 1. Note "wrap-around" of resource blocks, e.g., with respect to f₅Offset of 3 is equal to f₀。

Frequency hopping has not previously been used in Frequency Division Multiplexing (FDM) and OFDM systems employing variable bandwidth allocation. One difficulty in applying frequency hopping techniques to systems that allow variable bandwidth allocation is that the number of available hopping patterns varies according to the bandwidth allocation. For wideband signals, there are fewer hopping options than narrowband signals. For example, in an OFDM system with eight resource blocks in the frequency domain, there are eight different hopping possibilities (eight possible frequency locations) for the transmission bandwidth corresponding to one resource block. However, for a transmission bandwidth of seven resource blocks, there are only two hopping possibilities (two possible frequency locations). Therefore, the same hopping pattern cannot be used in both cases.

Furthermore, when mixing transmissions from two or more mobile devices using different bandwidths within one subframe (FDMA), the hopping probability of each mobile device depends on the bandwidth allocated to the other mobile devices. This constraint is shown in fig. 6. Fig. 6 shows two mobile terminals sharing a total of eight resource blocks. Mobile terminal 1 is allocated seven resource blocks while mobile terminal 2 is allocated only one resource block. From this simplified example it can be seen that there are only two possible frequency positions for the mobile terminal 1. Without other users, the mobile terminal 2 would have eight possibilities. However, to avoid collisions with the mobile terminal 1, the mobile terminal 2 is also limited to only two possible frequency locations.

A third problem is that bandwidth allocation is dependent on the instantaneous channel conditions of the mobile device and therefore cannot be known in advance. If the frequency pattern is established without regard to bandwidth allocation, the predetermined frequency hopping pattern will impose an undesirable constraint on the bandwidth allocation. In this case, bandwidth allocation must be made to avoid collisions, which will reduce the efficiency of the system.

The present invention provides a method for implementing frequency hopping in an OFDM system that allows variable bandwidth allocation. In accordance with the present invention, a scheduler at a base station or within a network dynamically determines both the bandwidth allocation and the frequency hopping pattern to be used by each mobile terminal scheduled during a given scheduling interval. The scheduling is therefore not based on a predefined hopping pattern. The scheduler then explicitly signals the bandwidth allocation and the frequency hopping pattern to the scheduled mobile terminals in a scheduling grant. In this way, the frequency hopping pattern can change from one scheduling interval to the next according to the bandwidth allocation.

FIG. 7 provides a simple example to illustrate how scheduling is performed according to an example embodiment. Fig. 7 shows one OFDM carrier with 24 resource blocks. In the following discussion, the index i denotes the mobile terminal, the index j denotes the slot, L_iBandwidth allocation for the ith mobile terminal in terms of number of resource blocks, and K_i(j) Is the frequency offset in the jth slot for the ith mobile terminal. Three mobile terminals are being scheduled for simultaneous transmission during a scheduling interval (e.g., a subframe) that includes two time slots. A first mobile terminal, denoted mobile terminal 1, is allocated eight resource blocks, a second mobile terminal, denoted mobile terminal 2, is allocated twelve resource blocks, and a third mobile terminal, denoted mobile terminal 3, is allocated 4 resource blocks. The bandwidth allocation is the same in each time slot during the scheduling interval. In the first time slot (time slot "0"), mobile terminal 1 is assigned a frequencyRate shift K₁(0) Mobile terminal 2 is assigned a frequency offset K12₂(0) Mobile terminal 3 is assigned a frequency offset K of 0₃(0) 20. In the second time slot (time slot "1"), the mobile terminal 1 is assigned a frequency offset K₁(1) Mobile terminal 2 is assigned a frequency offset K of 0₂(1) 12 and the mobile terminal 3 is assigned a frequency offset K₃(1)＝8。

As can be seen from the example shown in fig. 7, three parameters need to be signaled to each mobile terminal: bandwidth allocation L for scheduling interval_iFrequency offset K for the first time slot_i(0) And a frequency offset K for the second time slot_i(1). It should be noted that the frequency offset for the second time slot is not dependent on the frequency offset used in the first time slot, since no predefined hopping pattern is used. Thus, in the above example, the base station needs to signal the frequency offset for the second time slot as well as the first time slot. This process is referred to herein as explicit (explicit) signaling.

These three parameters L_i(allocated bandwidth in terms of number of resource blocks), K_i(0) (frequency offset for allocation of first time slot), and K_i(1) The frequency offset (for the allocation of the second time slot) may be signaled independently of each other. However, at L_iValue of and K_i(0) And K_i(1) There is a correlation between the possible values of (c). More precisely, for a given value L_i、K_i(0) And K_i(1) Can only take 0 to N-L_iWherein N is the total number of available resource blocks. Thus, by applying the parameter L_i、K_i(0) And K_i(1) Joint coding can be performed to reduce signaling L_i、K_i(0) And K_i(1) The total number of bits. This may be expressed such that L is signaled as a single parameter_i、K_i(0) And K_i(1) Instead of signalling L as three different independent parameters_i、K_i(0) And K_i(1)。

In some cases, frequency hopping may not be used all the time. One such scenario is when frequency domain channel dependent scheduling is used. If channel dependent scheduling is used, K_i(1) The explicit signaling of (2) implies unnecessary overhead. To avoid this situation, scheduling grants of different formats may be provided: comprising a parameter K_i(1) And a format not including the parameter K_i(1) The format of (a).

Fig. 8 illustrates an exemplary access node 50 for scheduling uplink transmissions in a mobile communication system. The access node 50 includes transceiver circuitry 52 coupled to an antenna 54 for communicating with one or more mobile terminals, and control circuitry 56 for controlling operation of the access node 50. The control circuitry 56 may include one or more processors that perform various control functions, such as radio resource control. The control circuit 56 includes a scheduler 58 to schedule uplink transmissions as described above. The scheduler 58 is responsible for determining which mobile terminals are scheduled for transmission during each scheduling interval and sending scheduling grants to the scheduled mobile terminals.

Fig. 9 illustrates an exemplary process 100 implemented by the scheduler 58. When frequency hopping is used, the process 100 shown in fig. 9 is repeated in each scheduling interval. Before the start of a given scheduling interval, the scheduler 58 selects a mobile terminal and determines a bandwidth allocation for the selected mobile terminal (block 102). The selection of the mobile terminal and the determination of the bandwidth allocation are based on channel conditions, buffer levels, and other relevant factors. Once the bandwidth allocation is determined, the scheduler 58 determines a frequency hopping pattern for each scheduled mobile terminal (block 104) and sends a scheduling grant to each scheduled mobile terminal (block 106).

The present invention provides a very flexible, simple (low complexity), and low overhead method to implement uplink frequency hopping in systems that support flexible bandwidth transmission. In general, those skilled in the art will recognize that the invention is not limited by the foregoing description and accompanying drawings. Instead, the present invention is limited only by the following claims and their legal equivalents.

Claims (16)

1. A method of scheduling transmissions in a mobile communication system, the method comprising:

determining a bandwidth allocation for at least one scheduled mobile terminal during a scheduling interval comprising two or more time slots;

determining a frequency hopping pattern for the scheduled mobile terminal over the scheduling interval based on the bandwidth allocation for the scheduled mobile terminal; and

transmitting the bandwidth allocation and frequency hopping pattern for the scheduling interval to the scheduled mobile terminals.

2. The method of claim 1, wherein determining a bandwidth allocation for a scheduled mobile terminal comprises: determining a number of subcarriers for the mobile terminal.

3. The method of claim 2, wherein determining the frequency hopping pattern of the scheduled mobile terminal over the scheduling interval comprises: determining frequency offsets for the mobile terminals for different time slots in the scheduling interval.

4. The method of claim 3, wherein mobile terminals are scheduled for uplink transmission, and wherein the bandwidth allocation and frequency hopping pattern for the scheduling interval to the mobile terminals comprises: transmitting the bandwidth allocation and frequency hopping pattern in a scheduling grant.

5. The method of claim 4, wherein transmitting a scheduling grant to the scheduled mobile terminal comprises: transmitting to the scheduled mobile terminal the number of subcarriers allocated to the scheduled mobile terminal and a set of frequency offsets for use by the scheduled mobile terminal in consecutive time slots of the scheduling interval.

6. The method of claim 5, wherein the number of subcarriers and the set of frequency offsets are transmitted as a single parameter.

7. The method of claim 1, wherein determining a frequency hopping pattern is further based on a bandwidth allocation for at least another simultaneously scheduled mobile terminal.

8. The method of claim 1, wherein determining the bandwidth allocation and determining the frequency hopping pattern comprises: a bandwidth allocation and a frequency hopping pattern are determined for two or more simultaneously scheduled mobile terminals in the same scheduling interval.

9. A scheduler for scheduling transmissions of a plurality of mobile devices in a mobile communication system, the scheduler being configured to:

determining a bandwidth allocation for at least one scheduled mobile terminal during a scheduling interval comprising two or more time slots;

determining a frequency hopping pattern for the scheduled mobile terminal over the scheduling interval based on the bandwidth allocation for the scheduled mobile terminal; and

transmitting the bandwidth allocation and frequency hopping pattern for the scheduling interval to the scheduled mobile terminals.

10. The scheduler of claim 9 configured to determine a bandwidth allocation for a scheduled mobile terminal by determining a number of subcarriers for the mobile terminal.

11. The scheduler of claim 10 configured to determine the frequency hopping pattern of the mobile terminal over the scheduling interval by determining frequency offsets for the scheduled mobile terminal for different time slots in the scheduling interval.

12. The scheduler of claim 11 wherein the scheduler is configured to schedule uplink transmissions from the mobile terminals, wherein the scheduler is configured to transmit the bandwidth allocations and frequency hopping patterns to the mobile terminals in a scheduling grant.

13. The scheduler of claim 12 configured to transmit to the scheduled mobile terminal the number of subcarriers allocated to the scheduled mobile terminal and a set of frequency offsets for use by the scheduled mobile terminal in consecutive time slots of the scheduling interval.

14. The scheduler of claim 13 wherein the number of subcarriers and the set of frequency offsets are transmitted by the scheduler as a single parameter.

15. The scheduler of claim 9 further configured to determine a frequency hopping pattern based on a bandwidth allocation for at least another simultaneously scheduled mobile terminal.

16. The scheduler of claim 9 configured to determine bandwidth allocations and frequency hopping patterns for two or more simultaneously scheduled mobile terminals in the same scheduling interval.