Method and apparatus for adaptive, variable bandwidth, high-speed data transmission of a multicarrier signal over digital subscriber lines

Claims

What is claimed is:

1. In a multicarrier data transmission system that transmits data continuously over dispersive, noisy subscriber loops for digital service from a telephone central office to a user location and corresponding digital service between transmitters and receivers at said user location and said central office, a method for implementing variable transmission bandwidth as a function of line frequency-variable attenuation, noise power spectrum, a set of programmed and possibly variable carrier target bit-error-rates, and a programmed and possibly frequency-variable transmit power mask to improve data transmission speed or to improve performance margin at any given transmission speed subject to a power or power-spectral-density constraint, said method using subcarrier-indexed measurements of channel gains, channel noises, desired carrier bit-error-rates, and the programmed power mask to provide subcarrier-indexed estimates of transmission quality and to maintain the transmission quality at high levels, comprising the steps of:

(a) sorting the subcarrier-indexed estimates of the transmission quality, scaled by the desired subcarrier bit-error-rates, into an invertible ordering for assessment of the relative data-carrying capabilities of the subcarriers at initialization and/or during data transmission;

(b) calculating bit and energy allocation tables for said multicarrier data transmission system based on the sorted subcarrier-indexed estimates for either improving aggregate transmitted data rate at fixed performance margin with said power or power-spectral-density constraint or improving performance margin at a fixed data rate with said power or power-spectral-density constraint;

(c) communicating said bit and energy allocation tables between the transmitters and the receivers of said multicarrier transmission system; and

(d) implementing said bit and energy allocation tables in coordination between the transmitters and the receivers during initialization of said multicarrier transmission system and/or during simultaneous transmission of said digital data by said multicarrier transmission system.

2. A multicarrier transmission system as recited in claim 1 and further comprising the steps of:

communicating changes in bit allocation by removing a single or several bits from those bits allocated to a carrier or subcarrier with a quality estimate below a specified threshold and placing that bit or bits on a second subcarrier or carriers with a quality estimate exceeding a second specified threshold; and

coordinating the implementation of said changes in both the transmitters and the receivers by communication through a bi-directional overhead data channel, said overhead data channel being simultaneously present with the user data channel on the same communication line.

3. A multicarrier transmission system as recited in claim 1 and further comprising the steps of:

communicating the computed bit and energy allocation tables with reliable means of error detection or correction by redundantly transmitting said allocation tables on several subcarriers from the receivers to the transmitters; and

coordinating the implementation of said allocation tables in both the transmitters and the receiver by communication through the communication line.

4. A multicarrier transmission system as recited in claim 1 and further comprising the steps of:

making subcarrier quality estimates through a computation of the average squared difference between an expected decision threshold on a subcarrier and the detected subcarrier signal level; and

comparing said quality estimates to thresholds for each subcarrier to determine whether or not a particular bit allocation or gain allocation change is to be implemented.

5. A multicarrier transmission system as recited in claim 1 and further comprising the steps of:

making subcarrier quality estimates through the computation of an average signal-to-noise measure on a subcarrier; and

comparing said quality estimates to thresholds for each subcarrier to determine whether or not a particular bit allocation or gain allocation change is to be implemented.

6. A multicarrier transmission system as recited in claim 1 and further comprising the steps of:

making subcarrier quality estimates through the computation of an average signal-to-noise ratio multiplied by a subcarrier dependent programmed reliability factor; and

comparing said quality estimates to thresholds for each subcarrier to determine whether or not a particular bit allocation or gain allocation change is to be implemented.

7. A multicarrier transmission system as recited in claim 2 and further comprising the steps of:

communicating the computed bit and energy allocation tables with reliable means of error detection or correction by redundantly transmitting said allocation tables from the receivers to the transmitters; and

coordinating the implementation of said allocation tables in both the transmitters and the receivers by communication through the communication line.

8. A multicarrier transmission system as recited in claim 1, 2, 3 or 7, and further comprising the steps of:

making subcarrier quality estimates through a computation of the average squared difference between an expected decision threshold on a subcarrier and the detected subcarrier signal level; and

comparing said quality estimates to thresholds for each subcarrier to determine whether or not a particular bit allocation or gain allocation change is to be implemented.

9. A multicarrier transmission system as recited in claim 1, 2, 3 or 7, and further comprising the steps of:

making subcarrier quality estimates through the computation of an average signal-to-noise measure on a subcarrier; and

comparing said quality estimates to thresholds for each subcarrier to determine whether or not a particular bit allocation or gain allocation change is to be implemented.

10. A multicarrier transmission system as recited in claim 1, 2, 3 or 7, and further comprising the steps of:

making subcarrier quality estimates through a computation of the average signal-to-noise ratio multiplied by a subcarrier dependent programmed reliability factor; and

comparing said quality estimates to thresholds for each subcarrier to determine whether or not a particular bit allocation or gain allocation change is to be implemented.

11. A multicarrier communication system comprising: a transmitter means and a receiver means communicatively linked together by a datalink, the signal carrying characteristics of which are randomly variable, said transmitter means including,

means for establishing a plurality of frequency domain subchannels for respectively communicating various quantities of data bits selected from a stream of data bits;

means for measuring the capability of the datalink to efficiently communicate the data bits in each said subchannel and for developing an optimum energy allocation and an optimum data bit allocation for each said subchannels; and

means for causing an input data stream to be divided and modulated onto a plurality of frequency domain carriers, each corresponding to one of said subchannels for transmission over said datalink to said receiver means, the quantity of data bits transmitted over each said subchannel being selected as a function of said optimum data bit allocation; and said receiver means including,

means for monitoring the quality of the transmission over each said subchannels; and

means for returning an indication of the monitored quality to said transmitter means.

12. A multicarrier communication system as recited in claim 11 wherein said transmitter means further includes: means responsive to said indication of monitored quality and operative to change the data bit allocation among said subchannels to improve the quality of the transmission.

13. A multicarrier communication system as recited in claim 11 wherein said means for measuring is initialized for maximizing total data rate in accordance with the following steps:

a) computing SNR(i) i, when all subcarriers are used and E(i)=1 i, where

SNR(i) is the SNR estimate of the i.sup.-th subcarrier at the receiver and

E (i) is the transmit energy of the i.sup.-th subcarrier, normalized to 1;

b) calculating SNR (i) /.GAMMA.(i) , where

.GAMMA.(i) is the "SNR gap" for the i.sup.-th subcarrier that is determined by the desired bit-error-rate, a system performance margin, a total effective coding gain, the Q-function, and the number of nearest neighbors in the input signal constellation for the i.sup.-th subcarrier;

c) sorting and storing the resulting SNR (i)/.GAMMA.(i)'s into descending order and labeling them so that SNR(j)/.GAMMA.(j)>SNR(j+1)/.GAMMA.(j+1) j<N;

d) letting k=1, b.sub.max =0, and initializing {b.sub.j } to all zeros, where

k is a count,

b.sub.max is the current maximum number of bits achievable in a multicarrier symbol, and

{b.sub.j } is the current bit allocation table that achieves b.sub.max ;

e) computing b.sub.target (k) according to ##EQU19## where b.sub.target (k) is the number of bits achievable in a multicarrier symbol using only the k best subcarriers, and

the number of bits achievable by the j.sup.-th subcarrier b.sub.j, is determined by: ##EQU20## wherein, round{x} is the integer rounding operation; i.e., when the decimal part of x.gtoreq.0.5, x is rounded up to the next larger integer value by the operation round{x} and when the decimal part of x<0.5, x is rounded down to the next smaller integer value no less than itself by the operation round{x},

floor{x} is the floor operation; i.e., x is always rounded down to the next smaller integer value no less than itself by the operation floor{x},

E.sub.target is the total input energy, or power, constraint at the transmitter,

k is the number of used subcarriers,

.GAMMA.(j) is the "SNR gap" of the chosen system parameters for the j.sup.-th subcarrier, and

E.sub.maxj is the maximum energy, or power, allowable in the j.sup.-th subcarrier due to the transmit power mask;

for each of the k subcarriers used, an equal amount of transmit energy E.sub.target /k is used to calculate the achievable number of bits per multicarrier symbol, but in no event is such energy greater than the allowable amount of individual carrier transmit energy determined by the power mask;

f) if b.sub.target (k)>b.sub.max, then setting b.sub.max =b.sub.target (k) and keeping track of the current bit allocation table, {b.sub.j }={b.sub.j }, that achieves b.sub.max ;

g) if k.noteq.N, then setting k=k+1 and returning to step e, otherwise going to step h; at this point b.sub.max will indicate the maximum achievable data rate, and the currently stored bit allocation table that achieves b.sub.max, {b.sub.j }, can be used to obtain the initial system bit allocation table, {b.sub.i 56 , since the mapping in step c is invertible;

h) allocating input energies, {E.sub.i }, accordingly so that P.sub.e (i)=P.sub.e,i,target i given the bit allocation table {b.sub.i }, where

P.sub.e (i) is the probability of error for the i.sup.-th used subcarrier and

P.sub.e,i,target is the target probability of error for the i.sup.-th subcarrier;

i) calculating the current total transmit energy, E.sub.total, according to ##EQU21## where E.sub.i is the current transmit energy used by the i.sup.-th subcarrier;

j) scaling final energy distribution,{E.sub.i }, according to ##EQU22## and k) causing said transmitter means to allocate bits to each said subchannel according to the bit allocation table {b.sub.i }) and to allocate transmitting power to each said subchannel in accordance with the distribution {E.sub.i }.

14. A multicarrier communication system as recited in claim 11 wherein said means for measuring is initialized for maximizing system performance margins subject to a predetermined target rate in accordance with the following steps:

a) computing SNR(i) i, when all subcarriers are used and E(i)=1 i, where

SNR(i) is the SNR estimate of the i.sup.-th subcarrier at the receiver, and

E(i) is the transmit energy of the i.sup.-th subcarrier, normalized to 1;

b) sorting and storing the resulting SNR(i)'s into descending order and labeling them so that SNR(j).gtoreq.SNR(j+1) j<N;

c) letting k=1, .gamma..sub.max =.infin., and count=0, where

k is a count,

.gamma..sub.max is the current maximum achievable system performance margin, and

count is the number of subcarriers to use to achieve .gamma..sub.max ;

d) computing .gamma.(k) according to ##EQU23## where .gamma.(k) is the maximum system performance margin achievable in a multicarrier symbol with a target data rate of b.sub.target, a total effective coding gain of .gamma..sub.eff, and a desired bit-error-rate of 10.sup.-7, using only the k best subcarriers, and

the current geometric SNR, SNR(k), is calculated as ##EQU24## where the current transmit energy used by the j.sup.-th subcarrier, E.sub.j, is determined by ##EQU25## wherein, E.sub.target is the total input energy, or power, constraint at the transmitter,

k is the number of used subcarriers, and

E.sub.maxj the maximum energy, or power, allowable in the j.sup.-th subcarrier due to the transmit power mask; for each of the k subcarriers used, an equal amount of transmit energy E.sub.target /k is used to calculate the current geometric SNR, SNR(k), but in no event is such energy greater than the allowable amount of individual carrier transmit energy due to the power mask;

e) if .gamma.(k)>.gamma..sub.max, then setting .gamma..sub.max =.gamma.(k) and count=k;

f) if k.noteq.N, then setting k=k+1 and returning to step d, otherwise going to step g; at this point .gamma..sub.max will indicate the maximum achievable system performance margin given the set of system parameters, and count will be the corresponding number of best subcarriers to use to achieve .gamma..sub.max ;

g) calculating the initial bit allocation table, {b.sub.i }, based on .gamma..sub.max and count according to ##EQU26## for those used subcarriers determined by count and the invertible mapping of step b, keeping track of the amount of rounding, diff.sub.j, by ##EQU27## and .GAMMA..sub.max is given by ##EQU28## using .gamma..sub.max =.gamma..sub.margin ; h) letting B.sub.total =.SIGMA..sub.i=0.sup.N-1 b.sub.i,

where B.sub.total is the total number of bits supported in a multicarrier symbol by the current bit allocation table {b.sub.i };

i) if B.sub.total >B.sub.target, then subtracting one bit at a time from the current bit table, {b.sub.i }, on the subcarrier that presently has the largest value of diff.sub.i, adjusting diff.sub.i =diff.sub.i -1 and B.sub.total =B.sub.total -1 respectively, and repeating until B.sub.total =B.sub.target ;

j) if B.sub.total <B.sub.target, then adding one bit at a time to the current bit table, {b.sub.i }, on the subcarrier that presently has the smallest value of diff.sub.i, adjusting diff.sub.i =diff.sub.i +1 and B.sub.total =B.sub.total +1 respectively, and repeating until B.sub.total =B.sub.target ;

k) allocating input energies, {E.sub.i }, accordingly so that P.sub.e (i)=P.sub.e,i,target i given the bit allocation table {b.sub.i }, where

P.sub.e (i) is the probability of error for the i.sup.-th used subcarrier, and

P.sub.e,i,target is the target probability of error for the i.sup.-th subcarrier;

l) calculating the current total transmit energy, E.sub.total, according to ##EQU29## where E.sub.i is the current transmit energy used by the i.sup.-th subcarrier;

m) scaling the final energy distribution,{E.sub.i }, according to ##EQU30## and n) causing said transmitter means to allocate said bits to each said subchannel according to the bit allocation table {b.sub.i } and to allocate transmitting power to each said subchannel in accordance with the distribution {E.sub.i }.

15. A multichannel communication system as recited in claim 12 wherein said transmitting means and said receiver means adaptively modify the system bit and/or transmission bandwidth during normal system operation in accordance with the following steps:

a) monitoring the steady state MSE's (or SNR's or SNR/.GAMMA.'s) of all used subcarriers in the background at said receiver means according to

.vertline.E.sub.ij .vertline..sup.2 =R{E.sub.ij.sup.2 }+I{E.sub.ij.sup.2 }

.epsilon..sub.ij =(.lambda.).epsilon..sub.ij-1 +(1-.lambda.).vertline.E.sub.ij .vertline..sup.2

where {E.sub.ij } is the error, or difference between the input and the output of the hard decision decoder at time j for the i.sup.-th subcarrier, and it is computed independently for the real (R{E.sub.ij }) and the imaginary (I{E.sub.ij }) dimensions for each subcarrier i; the cumulative error energy estimate for the i.sup.-th subcarrier at time j, .epsilon..sub.ij, is then computed by averaging successive squared error magnitudes, where .lambda. is a positive fraction slightly less than 1; SNR, or SNR/.GAMMA., normalization is included if that is the preferred transmission quality measure;

b) searching through all used subcarriers for .epsilon..sub.maxj and .epsilon..sub.minj in the background at said receiver means, where .epsilon..sub.maxj is the maximum cumulative error energy estimate and .epsilon..sub.minj is the minimum cumulative error energy estimate at time j, respectively;

c) if .epsilon..sub.maxj >.epsilon..sub.minj +.epsilon..sub.threshold then going to step d, otherwise returning to step a,

where .epsilon..sub.threshold is a predetermined threshold value that needs to be exceeded before a bit swap operation can take place;

d) sending bit swap information back to said transmitter means via the control channel;

e) synchronizing the change through a simple handshake procedure;

f) letting b(min)=b(min)+1 and b(max)=b(max)-1, where

b(min) is the current number of bits assigned in the bit allocation table to the subcarrier with a cumulative error energy estimate of .epsilon..sub.minj , and

b(max) is the current number of bits assigned in the bit allocation table to the subcarrier with a cumulative error energy estimate of .epsilon..sub.maxj ;

g) letting .epsilon..sub.minj =2.epsilon..sub.minj and .epsilon..sub.maxj =0.5.epsilon..sub.maxj ; and

h) adjusting the decoder slicer settings in said transmitter means for the two subcarriers involved in the bit swap operation accordingly.

16. In a multicarrier transmission system having transmitters and receivers to transmit and receive data continuously over communication lines, a method for implementing variable bandwidth data transmission with a plurality of subcarriers as a function of measured signal-to-noise ratios of the subcarriers, comprising the steps of:

calculating signal-to-noise factors for the subcarriers as functions of the measured signal-to-noise ratios of the subcarriers to a signal-to-noise gap approximation of the subcarriers, said gap approximation being a function of a system performance margin;

sorting the calculated signal-to-noise factors;

computing a plurality of bit allocation tables for the subcarriers using the first k subcarriers as provided in the order of the sorted factors and where k increases from one to the number of subcarriers;

determining bit allocation to the subcarriers by selecting from the plurality of bit allocation tables the table with the highest bit allocation to the subcarriers;

calculating energy allocation to each of the subcarriers according to its bit allocation; and

implementing said calculated bit allocation and energy allocation to the subcarriers in the transmission of data to optimize data transmission speed during initialization of said multicarrier transmission system and/or during simultaneous transmission of said data by said multicarrier transmission system.

17. In a multicarrier transmission system as recited in claim 16 wherein the calculating energy allocation step includes the substeps of:

calculating preliminary energy allocation to each of the subcarriers according to its bit allocation such that a probability-of-error for each of the subcarriers equals to a target probability of error for each of the subcarriers;

determining the energy allocation to each of the subcarriers by finding the smaller of a proportional preliminary energy allocation value and a maximum subcarrier energy value, said proportional preliminary energy allocation value being a function of the total preliminary energy allocation to the subcarriers, an energy constraint,,and the preliminary energy allocation to each of the subcarriers.

18. In a multicarrier transmission system as recited in claim 17 further including the steps of:

measuring subcarrier transmission quality estimates through the computation of an average squared difference between an expected decision threshold on a subcarrier and the detected subcarrier signal level; and

comparing said quality estimates to threshold values for the subcarriers to determine whether bit swapping from the subcarriers with worse quality estimates to subcarriers with better quality estimates is to be implemented.

19. In a multicarrier transmission system as recited in claim 17 further including the steps of:

measuring subcarrier transmission quality estimates through the computation of the average signal-to-noise ratios on the subcarriers; and

comparing said quality estimates to threshold values for the subcarriers to determine whether bit swapping from the subcarriers with worse quality estimates to subcarriers with better quality estimates is to be implemented.

20. In a multicarrier transmission system as recited in claim 17 further including the steps of:

measuring subcarrier transmission quality estimates through the computation of the average signal-to-noise ratios multiplied by a subcarrier dependent programmed reliability factor; and

comparing said quality estimates to threshold values for the subcarriers to determine whether bit swapping from the subcarriers with worse quality estimates to subcarriers with better quality estimates is to be implemented.

21. In a multicarrier transmission system as recited in claim 16 wherein the calculating energy allocation step includes the substeps of:

calculating preliminary energy allocation to each of the subcarriers according to its bit allocation such that a probability-of-error for each of the subcarriers equals to a target probability of error for each of the subcarriers; and

determining the energy allocation to each of the subcarriers by finding the smaller of a proportional preliminary energy allocation value and a maximum subcarrier energy value, said proportional preliminary energy allocation value being a function of the total preliminary energy allocation to the subcarriers, an energy constraint, and the preliminary energy allocation to each of the subcarriers.

22. In a multicarrier transmission system as recited in claim 16 further including the steps of:

measuring subcarrier transmission quality estimates through the computation of an average squared difference between an expected decision threshold on a subcarrier and the detected subcarrier signal level; and

comparing said quality estimates to threshold values for the subcarriers to determine whether bit swapping from the subcarriers with worse quality estimates to subcarriers with better quality estimates is to be implemented.

23. In a multicarrier transmission system as recited in claim 16 further including the steps of:

measuring subcarrier transmission quality estimates through the computation of the average signal-to-noise ratios on the subcarriers; and

comparing said quality estimates to threshold values for the subcarriers to determine whether bit swapping from the subcarriers with worse quality estimates to subcarriers with better quality estimates is to be implemented.

24. In a multicarrier transmission system as recited in claim 16 further including the steps of:

measuring subcarrier transmission quality estimates through the computation of the average signal-to-noise ratios multiplied by a subcarrier dependent programmed reliability factor; and

comparing said quality estimates to threshold values for the subcarriers to determine whether bit swapping from the subcarriers with worse quality estimates to subcarriers with better quality estimates is to be implemented.

25. In a multicarrier transmission system having transmitters and receivers to transmit and receive data continuously over communication lines, a method for implementing variable bandwidth data transmission with a plurality of subcarriers and as a function of measured signal-to-noise ratios of the subcarriers, comprising the steps of:

sorting the signal-to-noise ratios of the subcarriers;

computing preliminary system performance margins using the subcarriers in the order of the sorted ratios, the preliminary subsystem performance margins being a function of the signal-to-noise ratios, the number of subcarriers, a target data rate, a total effective coding gain, a desired bit-error-rate, and energy for the subcarriers;

selecting the maximum system performance margin from said computed preliminary system performance margins;

calculating bit allocations to the subcarriers using said maximum system performance margin;

calculating energy allocation to the subcarriers according to its bit allocation; and

implementing said calculated bit allocation and energy allocation to the subcarriers in the transmission of data to optimize data transmission speed during initialization of said multicarrier transmission system and/or during simultaneous transmission of said data by said multicarrier transmission system.

26. In a multicarrier transmission system as recited in claim 25 wherein the calculating bit allocation step includes the substeps of:

calculating a preliminary bit allocation table using the system performance margin; and

determining the bit allocations to the subcarriers from the preliminary bit allocation table by comparing the total number of bits in the preliminary bit allocating table with the target data rate to determine whether bits should be added or removed from the preliminary bit allocation table.

27. In a multicarrier transmission system as recited in claim 26 wherein the calculating energy allocation step includes the substeps of:

calculating preliminary energy allocation to each of the subcarriers according to its bit allocation such that a probability-of-error for each of the subcarriers equals to a target probability of error for each of the subcarriers;

determining the energy allocation to each of the subcarriers by finding the smaller of a proportional preliminary energy allocation value and a maximum subcarrier energy value, said proportional preliminary energy allocation value being a function of the total preliminary energy allocation to the subcarriers, an energy constraint, and the preliminary energy allocation to each of the subcarriers.

28. In a multicarrier transmission system as recited in claim 27 further including the steps of:

measuring subcarrier transmission quality estimates through the computation of an average squared difference between an expected decision threshold on a subcarrier and the detected subcarrier signal level; and

comparing said quality estimates to threshold values for the subcarriers to determine whether bit swapping from the subcarriers with worse quality estimates to subcarriers with better quality estimates is to be implemented.

29. In a multicarrier transmission system as recited in claim 27 further including the steps of:

measuring subcarrier transmission quality estimates through the computation of the average signal-to-noise ratios on the subcarriers; and

comparing said quality estimates to threshold values for the subcarriers to determine whether bit swapping from the subcarriers with worse quality estimates to subcarriers with better quality estimates is to be implemented.

30. In a multicarrier transmission system as recited in claim 27 further including the steps of:

measuring subcarrier transmission quality estimates through the computation of the average signal-to-noise ratios multiplied by a subcarrier dependent programmed reliability factor; and

comparing said quality estimates to threshold values for the subcarriers to determine whether bit swapping from the subcarriers with worse quality estimates to subcarriers with better quality estimates is to be implemented.

31. In a multicarrier transmission system as recited in claim 25 wherein the calculating bit allocation step includes the substeps of:

calculating a preliminary bit allocation table using the system performance margin; and

determining the bit allocations to the subcarriers from the preliminary bit allocation table by comparing the total number of bits in the preliminary bit allocating table with the target data rate to determine whether bits should be added or removed from the preliminary bit allocation table.

32. In a multicarrier transmission system as recited in claim 25 wherein the calculating energy allocation step includes the substeps of:

calculating preliminary energy allocation to each of the subcarriers according to its bit allocation such that a probability-of-error for each of the subcarriers equals to a target probability of error for each of the subcarriers; and

determining the energy allocation to each of the subcarriers by finding the smaller of a proportional preliminary energy allocation value and a maximum subcarrier energy value, said proportional preliminary energy allocation value being a function of the total preliminary energy allocation to the subcarriers, an energy constraint, and the preliminary energy allocation to each of the subcarriers.

33. In a multicarrier transmission system as recited in claim 25 further including the steps of:

measuring subcarrier transmission quality estimates through the computation of an average squared difference between an expected decision threshold on a subcarrier and the detected subcarrier signal level; and

comparing said quality estimates to threshold values for the subcarriers to determine whether bit swapping from the subcarriers with worse quality estimates to subcarriers with better quality estimates is to be implemented.

34. In a multicarrier transmission system as recited in claim 25 further including the steps of:

measuring subcarrier transmission quality estimates through the computation of the average signal-to-noise ratios on the subcarriers; and

comparing said quality estimates to threshold values for the subcarriers to determine whether bit swapping from the subcarriers with worse quality estimates to subcarriers with better quality estimates is to be implemented.

35. In a multicarrier transmission system as recited in claim 25 further including the steps of:

measuring subcarrier transmission quality estimates through the computation of the average signal-to-noise ratios multiplied by a subcarrier dependent programmed reliability factor; and

comparing said quality estimates to threshold values for the subcarriers to determine whether bit swapping from the subcarriers with worse quality estimates to subcarriers with better quality estimates is to be implemented.