Method for adaptively allocating resources in communication system

Abstract

A method for adaptively allocating resource can be simply implemented for reducing degradation of performance by effectively separating operations of sub-channel and time slot allocation and selection of modulation method and sequentially processing each of operations. A method for adaptively allocating resource in a communication system to sequentially process sub-carrier/time slot allocation and modulation method selection efficiently includes the step of a) computing average channel gains of sub-carriers/time slots for each user by using channel gains of sub-carriers/time slots for each user; b) computing average numbers of bits for each user by using required data rates and average channel gains of sub-carriers/time slots for each user; c) computing the number of sub-carriers/time slots allocated to each user and allocating the sub-carriers/time slots to each user; and d) selecting a modulation method with respect to each sub-carrier/time slot.

Description

In the equation 3, N₀/2 is a variance of white gauss noise, p_e is a bit error rate, and Q(x) is a Q function.

The equation 3 is a convex function, which can be applied to QAM, PSK, and PAM.

If a non-linear equation with K+1 equations and variables of the equation 2 is a convex function, there always exist an optimal solution. Therefore, c_k can be obtained by a Newton method disclosed by K. E. Atkinson, Numerical Analysis, Wiley & Sons Inc., 1998.

At Step S307, a total number of sub-carriers for each user is computed by equation 4. A total number of bits is R_k in below equation 4.
n_k=R_k√{square root over (c_k)}, for k=1, . . . , K   [Equation 4]

If the average number of bits c_k from the equation 2 and the number of sub-carriers n_k of each user from the equation 4 are substituted into the optimal solution, a simplified optimal solution is obtained as equation 5.

$\begin{array}{cc}\mathrm{Minimize}{P}_i-\sum _{k=1}^K\sum _{n=1}^N{r}_{k,n}{\rho }_{k,n}\mathrm{Subject}\mathrm{to}\sum _{n=1}^N{\rho }_{k,n}=n_k,\mathrm{for}\mathrm{all}k\sum _{k=1}^K{\rho }_{k,n}=1,\mathrm{for}\mathrm{all}n& \left[\mathrm{Equation}5\right]\end{array}$

In the equation 5, ρ_k,n is a binary variable, which shows if the K-th user uses the n-th sub-carrier. The value is 1 for using and 0 for not using.

That is, at step S311, the sub-carriers are allocated by solving the optimal solution of the equation 5.

An r_k,n of the equation 5 is the cost for the K-th user to use the n-th sub-carrier, and this relationship is further described in the below equation 6.
r_k,n=ƒ( c_k)/α²_k,n, for k=1 , . . . , K and n=1, . . . , N   [Equation 6]

Equation 5 is basically an integer type optimal solution because of the binary variable ρ_k,n. Although the computation is complicated to obtain the integer type optimal solution, the equation 5 is a particular solution, which can be solved with the integer condition of ρ_k,n removed.

Although the optimal solution can be solved with a Simplex method, it can not be practically implemented in real time.

When the optimal solution of the equation 5 is reviewed for minimizing the complexity, it is regarded as a transportation problem which is a particular solution of a linear optimal problem. In this case, N sub-carriers are suppliers and K users are consumers. The first constraint is that each consumer demands n_k items. The second constraint is that all suppliers supply only one item.

This type of transportation problem becomes a very simple computation with Vogel's method. The solution has small performance degradation compared to the optimal solution. The Simplex Method and the Vogel's Method are disclosed by W. L. Winston, entitled Operations Research, Duxbury Press, 1994.

At Step S311, the sub-carriers are allocated to each user by the equation 5 and at Step S313, a modulation method for each user is selected.

Once the sub-carrier allocation is finished, a condition that a plurality of users transmits data on one sub-carrier is removed. Then, bits are allocated on the allocated channel of each user by using the modulation method for an individual OFDM user.

The sub-carrier allocation and the modulation method selection are described in the preferred embodiment of the present invention. Those skilled in the art will also obviously find out that the allocation and the modulation method selection for time slots are performed in the same manner as those for the sub-carrier.

FIGS. 4 to 6 are graphs for describing the preferred embodiment of the present invention, wherein there are 4 users, 64 sub-carriers, and 8 time slots.

FIGS. 4A to 4D are graphs showing channel responses of user in accordance with another preferred embodiment of the present invention.

Each user has a different average channel gain in accordance with a distance estimated from the base station. The average channel gains of each user are shown as 0.0025, 0.3922, 1.3452 and 2.2601 in the preferred embodiment of the present invention.

As shown in FIGS. 4A to 4D, the sub-carriers have different values according to each user, but the values do not vary with the time slots. Therefore, the channel response of the first channel for user 1 is the same value with respect to the time slots 1 to 8. Total numbers of allocated channels for each of the 4 users computed from the equation 1 to 4 are 297, 83, 67, 62, respectively. The total number of channels is multiple of the number of sub-carriers and the number of time slots, which is 512.

The user 1 has a relatively large number 297 of the sub-carriers because the user 1 has a low average channel response 0.0025. The user 4 has the least number 62 of the sub-carriers because the user 4 has the highest average channel response 2.2601.

The optimal solution of the equation 5 can be solved with these numbers of the allocated sub-carriers of each user.

FIG. 5 is an exemplary drawing showing results of sub-carrier allocations in accordance with the present invention.

The dark mark shows the allocated channel to the user and the white mark shows the channel which is not allocated to the user.

FIG. 6 is an exemplary drawing showing results of users selected different methods of modulation in accordance with the present invention.

The white mark represents the channel which bits are not allocated while the dark mark represents the channel which bits are allocated. Particularly, the darker marks, the more bits are allocated in the channel. QPSK is 2 bits transmission and 16QAM is 4 bits transmission.

Comparing FIGS. 5 and 6, the bits are allocated on each user's allocated sub-carrier in accordance with a magnitude of the sub-carriers. The users having a large average channel gain are allocated with the small number of sub-carriers, which transmit the large number of bits. The user 1 is allocated with the large number of sub-carriers and using low order modulation methods because of a poor channel response.

All users have channels which may not have any bit during the modulation method selection according to the channel magnitude as shown in FIG. 5.

Table 1 shows a performance difference between the optimal solution proposed by the prior art disclosed in the previously mentioned article and the present invention with 4 users and 64 sub-carriers.

TABLE 1
		Power of the
Data Rate	Power of the	suboptimal solution
(bits/OFDM	optimal solution	of the present
symbol)	of the prior art	invention
128	27.62 dB	27.54 dB
256	34.49 dB	34.41 dB
384	40.56 dB	40.50 dB

Table 1 is an averaged result of 1000 times performed trials to get an average performance difference.

It is assumed that required data rates of each user are identical. The total data rate is varied from 128 bits/OFDM symbol to 384 bits/OFDM symbol. In case of 128 bits/OFDM symbol, QPSK is used without the adaptive modulation because the number of sub-carriers is 64.

The present invention offers significant simplicity compared to the optimal solution of the prior art while incurring small performance degradation of 0.6˜0.8 dB according to Table 1.

As a result, the present invention can execute the adaptive sub-carrier/time-slot allocation and the modulation method selection with the computation, which is practical to be implemented in the hybrid OFDMA/TDMA system.

Also, the present invention obtains significant power gain compared to the prior fixed modulation method and increase the efficiency of frequency usage.

The method of the present invention can be implemented as a program and stored in computer readable medium, e.g., a CD-ROM, a RAM, a ROM, a floppy disk, a hard disk and an optical/magnetic disk.

The present invention can efficiently execute allocation of sub-carriers and time slots when the hybrid OFDMA/TDMA is used as multi accessing method in a data communication system that is operated in the OFDM.

Also, the present invention obtains more power gain and is more efficient in using frequencies than the conventional system that uses the fixed modulation method.

While the present invention has been described with respect to certain preferred embodiments, it will be apparent to those skilled in the art that various changes and modifications may be made without departing from the scope of the invention as defined in the following claims.

Claims

1. A method for adaptively allocating resources in a communication system by subsequently performing sub-carrier/time slot allocation and modulation selection processes, the method comprising the steps of:

a) computing average channel gains of sub-carriers/time slots for each user based on channel gains of sub-carriers/time slots for each user;

b a) computing average numbers of bits of sub-carriers/time slots for each user based on required data rates and the computed an average channel gains gain of sub-carriers/time slots for each user;

c b) computing the number of sub-carriers/time slots to be allocated to each user based on the computed average numbers of bits of the sub-carriers/time slots for each user and allocating the sub-carriers/time slots to each user based on the computed average numbers of bits of the sub-carriers/time slots for each user and the computed number of sub-carriers/time slots to be allocated to each user;

d c) selecting a modulation process to modulate data according to a magnitude of the sub-carrier/time slot allocated to each user; and

e d) modulating data to be transmitted for each user through the selected modulation process, and transmitting said modulated data,

wherein the average channel gain of sub-carriers/time slots for each user is computed based on channel gains of sub-carriers/time slots for each use.

2. The method as recited in claim 1, wherein the average channel gain of each user in the step a) is computed by using an equation as:

3. The method as recited in claim 1, wherein the average number of bits for each user in the step b a) is a solution of K+1 non-linear equations formulated by an equation as:

4. The method as recited in claim 1, wherein the number of sub-carriers/time slots in the step c b) is computed by using an equation as:

n_k=R_k√{square root over (c_k)}, for k=1, . . . , K

wherein, n_k is the number of allocated sub-carriers/time slots for each user.

5. The method as recited in claim 1, wherein the allocation of sub-carrier/time slot in the step c b) is an optimal solution of an equation as:

6. The method as recited in claim 5, wherein the cost for the K-th user to use the n-th sub-carrier is determined by an equation as:

r_k,n=ƒ( c_k)/α²_k,n,

for

k=1, . . . , K and n=1, . . . , N.

7. The method as recited in claim 5, wherein a linear optimal solution is solved by applying a Vogel's method.

8. A non-transitory computer readable recording medium for storing programs for executing a method for adaptively allocating resources in a communication system including a microprocessor by subsequently performing sub-carrier/time slot allocation and modulation selection processes, comprising the steps of:

a) computing average channel gains of sub-carriers/time slots for each user based on channel gains of sub-carriers/time slots for each user;

b a) computing average numbers of bits of sub-carriers/time slots for each user based on required data rates and the computed average channel gains of sub-carriers/time slots for each user;

c b) computing the number of sub-carriers/time slots to be allocated to each user based on the computed average numbers of bits of the sub-carriers/time slots for each user and allocating the sub-carriers/time slots to each user based on the computed average numbers of bits of the sub-carriers/time slots for each user and the computed number of sub-carriers/time slots to be allocated to each user;

d c) selecting a modulation process to modulate data according to a magnitude of the sub-carrier/time slot allocated to each user; and

e d) modulating data to be transmitted for each user through the selected modulation process, and transmitting said modulated data,

wherein the average channel gain of sub-carriers/time slots for each user is computed based on channel gains of sub-carriers/time slots for each user.

9. A method for adaptively allocating resources in a communication system, the method comprising:

computing an average number of bits of sub-carriers/time slots for each user based on an average channel gain of sub-carriers/time slots for each user;

computing a number of sub-carriers/time slots to be allocated to each user based on the computed average number of bits;

allocating the sub-carriers/time slots to each user based on the computed average number of bits and the computed number of sub-carriers/time slots;

selecting a modulation process to modulate data based on the sub-carrier/time slots allocated to each user;

modulating data to be transmitted for each user with the selected modulation process; and

transmitting the modulated data,

wherein the average channel gain of sub-carriers/time slots for each user is computed based on channel gains of sub-carrier/time slots for each user.

10. A non-transitory computer readable recording medium for storing programs for executing a method for adaptively allocating resources in a communication system including a microprocessor, the method comprising:

computing an average number of bits of sub-carriers/time slots for each user based on an average channel gain of sub-carriers/time slots for each user;

computing a number of sub-carriers/time slots to be allocated to each user based on the computed average number of bits;

allocating the sub-carriers/time slots to each user based on the computed average number of bits and the computed number of sub-carriers/time slots;

selecting a modulation process to modulate data based on the sub-carriers/time slots allocated to each user;

modulating data to be transmitted for each user with the selected modulation process; and

transmitting the modulated data, wherein the average channel gain of sub-carriers/time slots for each user is computed based on channel gains of sub-carriers/time slots for each user.

11. A base station having a transmitter, the transmitter comprising:

a sub-carrier allocation and modulation method selection unit configured to

compute an average number of bits of sub-carriers/time slots for each user based on an average channel gain of sub-carriers/time slots for each user,

compute a number of sub-carriers/time slots to be allocated to each user based on the computed average number of bits,

allocate the sub-carriers/time slots to each user based on the computed average number of bits and the computed number of sub-carriers/time slots, and

select a modulation process to modulate data based on the sub-carriers/time slots allocated to each user; and

an adaptive modulator coupled to the sub-carrier allocation and modulation method selection unit, wherein the adaptive modulator is configured to modulate data to be transmitted for each user with the selected modulation process,

wherein the average channel gain of sub-carriers/time slots for each user is computed based on channel gains of sub-carriers/time slots for each user.