CN103354531B - Bit loading method in a kind of Digital Subscriber Line up-link - Google Patents

Abstract

The invention discloses the bit loading method in a kind of Digital Subscriber Line up-link, its processing procedure is: according to the channel status of each user on all upstream subchannels all users are divided into weak user group and strong user organizes two groups, modulator-demodulator corresponding to each weak user adopts RA-LC bit loading method to carry out bit distribution to all upstream subchannels, modulator-demodulator corresponding to each strong user adopts the bit loading method based on power weightings to carry out bit distribution to all upstream subchannels, advantage is by utilizing strong user to the corresponding relation of the power division of the interference effect of weak user and weak user, the power increment of the power of weak user and strong user is weighted on average by strong user, then all choosing the minimum upstream subchannels of weighted power carries out bit loading at every turn, can more bit be loaded on high frequency subchannel like this, thus can effectively suppress strong user to the cross-talk of weak user, the message transmission rate of weak user can also be maximized.

Description

Bit loading method in a kind of Digital Subscriber Line up-link

Technical field

The present invention relates to the spectrum management techniques in a kind of cable broadband Access Network, especially relate to the bit loading method in a kind of Digital Subscriber Line up-link.

Background technology

The demand of modern society to information is increasing, and the rise of the business such as particularly video request program, many people video conference, makes people propose higher level requirement to communication service and means of communication.The disposal ability of the backbone bandwidth of Current Communication Network, the speed of router and server can meet various broadband services demand.The greatest difficulty setting up broadband communication network is the transmission rate how improving user access networks.

Second high-speed digital user wire (the SecondGenerationVery-high-bit-rateDSL of current extensive concern, VDSL2) access technology, will by spread-spectrum to 30MHz, and with the Digital Subscriber Line (DigitalSubscriberLine of other type, DSL) access technology is compatible, intend adopting the planning of advanced signal transacting, Nonlinear Dynamic and coding theory, 1km distance provides the speed of 30 ~ 50Mbps, and at 300m apart from the interior high-speed data transfer that can realize 100Mbps bi-directional symmetrical in theory.Therefore, second high-speed digital user wire access technology is considered to one of effective way eliminating last one kilometer bottleneck, is following 10 ~ 20 years global topmost broadband access methods.

Second high-speed digital user wire access technology carries out information transmission owing to adopting higher frequency range, make electromagnetic interference increase between twisted-pair feeder in same cable, this electromagnetic interference is exactly cross-talk (Crosstalk), cross-talk 15 ~ 20dB stronger than other types noise, become the principal element affecting second high-speed digital user wire access technology transmission performance, therefore how fast and effeciently to suppress cross-talk to be the key ensureing second high-speed digital user wire high-speed transfer.

Dynamic Spectrum Management (DynamicSpectrumManagement, DSM) be one of effective way reducing cross-talk, it can load two kinds of modes by Direct Power distribution and bit and realize, RA-LC(Rate-adaptiveLevin-Campello) be a kind of representative bit loading method, but each user maximizes self message transmission rate and the cross-talk ignored other users in RA-LC bit loading method, weak user is caused to be subject to the severe jamming of strong user thus, thus cause the message transmission rate of weak user very low, even cannot work.

Summary of the invention

Technical problem to be solved by this invention is to provide the bit loading method in a kind of Digital Subscriber Line up-link, and it effectively can suppress cross-talk, thus improves the uplink transmission rate of weak user in digital subscriber line system.

The present invention solves the problems of the technologies described above adopted technical scheme: the bit loading method in a kind of Digital Subscriber Line up-link, it is characterized in that its processing procedure is: all users are divided into weak user according to the channel status of each user on all upstream subchannels and organize U1 and strong user organizes U2 two groups; Weak user organizes modulator-demodulator corresponding to each weak user in U1 and adopts RA-LC bit loading method to carry out bit distribution to all upstream subchannels; Strong user organizes modulator-demodulator corresponding to each strong user in U2 and adopts the bit loading method based on power weightings to carry out bit distribution to all upstream subchannels, and the main process of distribution is: first, calculates the weighted factor of strong user; Then obtain the weighted power of strong user on each upstream subchannels, find out the weighted power that value is minimum; Then, according to the target data transmission rate of strong user, the weighted power of strong user on the upstream subchannels being worth minimum weighted power, the maximum power of strong user on the upstream subchannels being worth minimum weighted power limits and the up gross power of strong user limits, and carries out bit distribution and upgrade weighted factor; Finally, according to the difference of the up gross power restriction of strong user and the maximum weighted Summing Factor minimum weight factor of strong user, end bit assigning process is determined whether.

Bit loading method in Digital Subscriber Line up-link of the present invention, it specifically comprises the following steps:

1. digital subscriber line system adopts Discrete Multitone modulation, whole channel is divided into multiple independently subchannel, again uplink and downlink division is carried out to all subchannels, will the subchannel of uplink information transmission be used for as upstream subchannels, will the subchannel of downlink information transmission be used for as downlink sub-channels;

2. the Spectrum Management Center in Digital Subscriber Line up-link is according to the channel status of each user on all upstream subchannels, all users are divided into weak user and organize U1 and strong user organizes U2 two groups, then user grouping information is sent to each modulator-demodulator in Digital Subscriber Line up-link by Spectrum Management Center;

3., in Digital Subscriber Line up-link, a spectrum management model is built:

$\underset{s^1,...,s^n,...,s^N}{\max }\underset{n\∈U1}{\mathrm{\Σ}}{R}^n$

R ⁿ≥R ^n,target,n∈U2

$0\≤s_k^n\≤s_k^{n,\mathrm{mask}},n=1,...,N,k=1,...,K,$ This spectrum management model description is reach given target strong user

$\underset{k=1}{\overset{K}{\mathrm{\Σ}}}{s}_k^n\≤P^{n,\mathrm{tot}},n=1,...,N$

When data transmission rates demands, maximize the message transmission rate of weak user, wherein, N represents number of users, the number of the upstream subchannels in K representative digit subscribers feeder up-link, max () for getting max function, S ¹represent the power allocation vector of user 1 on all upstream subchannels, S ⁿrepresent the power allocation vector of user n on all upstream subchannels, s ⁿrepresent the power allocation vector of user N on all upstream subchannels, represent the power of user n on the 1st upstream subchannels, represent the power of user n on a kth upstream subchannels, represent the power of user n on K upstream subchannels, U1 represents weak user's group, R ⁿrepresent the message transmission rate of user n, R ^{n, target}represent the target data transmission rate of user n, U2 represents strong user's group, represent the maximum power restriction of user n on a kth upstream subchannels, P ^{n, tot}represent the up gross power restriction of user n;

4., in Digital Subscriber Line up-link, weak user organizes modulator-demodulator corresponding to each weak user in U1 and adopts RA-LC bit loading method to carry out bit distribution to all upstream subchannels;

At this, weak user organizes modulator-demodulator corresponding to any one weak user in U1 and adopts RA-LC bit loading method to the detailed process that all upstream subchannels carry out bit distribution to be:

-1 4., suppose that this weak user is user n, and obtained the power increment of user n on each upstream subchannels, suppose that user n is loaded with on a kth upstream subchannels individual bit, then load on a kth upstream subchannels by user n power increment required for individual bit is designated as wherein, n=1,2 ..., N, N represent number of users, k=1,2 ..., the number of the upstream subchannels in K, K representative digit subscribers feeder up-link, represent the bit number that user n loads on a kth upstream subchannels, order initial value be 0, represent that user n loads on a kth upstream subchannels power required for individual bit, $p_{\left(b_k^n\right)}=\mathrm{\Γ}(2^{b_k^n}-1)\frac{\underset{m\≠n}{\mathrm{\Σ}}{\left|h_k^{n,m}\right|}^2{s}_k^n+{\mathrm{\σ}}_k^n}{{\left|h_k^{n,n}\right|}^2},$ Γ is signal to noise ratio difference, represent that user m is to the cross-talk coefficient of user n on a kth upstream subchannels, represent the power of user m on a kth upstream subchannels, represent the thermal noise that user n is subject on a kth upstream subchannels, represent the fading coefficients of user n on a kth upstream subchannels, at this symbol " || " for plural number asks modulo operation symbol, represent that user n loads on a kth upstream subchannels power required for individual bit;

4.-2, from the power increment of user n on all upstream subchannels, the minimum power increment of value is found out;

4.-3, judge whether user n can cause the gross power of user n to exceed the up gross power restriction of user n load a bit on the upstream subchannels of the minimum power increment of this value after, if meeting, the modulator-demodulator that then user n is corresponding terminates the bit allocation procedures of all upstream subchannels, otherwise, continue to perform step 4.-4;

4.-4, judge whether to cause after user n loads a bit on the upstream subchannels of the minimum power increment of this value the power of user n on the upstream subchannels of the minimum power increment of this value to exceed the maximum power restriction of user n on the upstream subchannels of the minimum power increment of this value, if meeting, then the power increment of user n on the upstream subchannels of the minimum power increment of this value is set to infinity, then step 4.-2 continuation execution are returned, otherwise, user n loads a bit on the upstream subchannels of the minimum power increment of this value, then the power increment of user n on the upstream subchannels of the minimum power increment of this value is upgraded, then step 4.-2 continuation execution are returned,

5. fabricate a Virtual User r to represent all weak users, the power of Virtual User r on each upstream subchannels is determined by all power of weak user on each upstream subchannels, is designated as by the power of Virtual User r on a kth upstream subchannels wherein, k=1 ..., the number of the upstream subchannels in K, K representative digit subscribers feeder up-link, n=1,2 ..., N, N represent number of users, represent the power of user n on a kth upstream subchannels, d represents that weak user organizes the number of the weak user in U1;

The power allocation case of Virtual User r on each upstream subchannels sends to strong user to organize each strong user in U2 by the Spectrum Management Center 6. in Digital Subscriber Line up-link, and the modulator-demodulator that strong user organizes each strong user in U2 corresponding adopts the bit loading method based on power weightings to carry out bit distribution to all upstream subchannels;

At this, strong user organizes modulator-demodulator corresponding to any one strong user in U2 and adopts the bit loading method based on power weightings to the detailed process that all upstream subchannels carry out bit distribution to be:

-1 6., suppose that this strong user is user n, order represent the minimum weight factor of user n, and initial value be 0, order represent the maximum weighted factor of user n, and initial value be 1, make the power of user n on a kth upstream subchannels initial value be 0, k=1,2 ..., K;

6.-2, basis with calculate the weighted factor of user n, be designated as w ⁿ,

6.-3, judge $|\underset{k}{\mathrm{\Σ}}{s}_k^n-P^{n,\mathrm{tot}}|>P^{n.\mathrm{tot}}\×0.005$ With $w_{\max }^n-w_{\min }^n>0.0001$ Whether set up simultaneously, if set up, then perform step 6.-4, otherwise the modulator-demodulator that user n is corresponding terminates the bit allocation procedures of all upstream subchannels, wherein, P ^{n, tot}representing the up gross power restriction of user n, is the symbol that takes absolute value at this symbol " || ";

-4 6., obtain the weighted power of user n on each upstream subchannels, the weighted power of user n on kth sub-channels is designated as ${\mathrm{eff}}_k^n=w^n{\mathrm{\Δp}}_{(b_k^n+1)}+(1-w^n)s_k^r,$ Wherein, represent that user n loads the on a kth upstream subchannels power increment required for individual bit, n=1,2 ..., N, N represent number of users, k=1,2 ..., the number of the upstream subchannels in K, K representative digit subscribers feeder up-link, represent the bit number that user n loads on a kth upstream subchannels, order initial value be 0, represent that user n loads on a kth upstream subchannels power required for individual bit, $p_{\left(b_k^n\right)}=\mathrm{\Γ}(2^{b_k^n}-1)\frac{\underset{m\≠n}{\mathrm{\Σ}}{\left|h_k^{n,m}\right|}^2{s}_k^n+{\mathrm{\σ}}_k^n}{{\left|h_k^{n,n}\right|}^2},$ Γ is signal to noise ratio difference, represent that user m is to the cross-talk coefficient of user n on a kth upstream subchannels, represent the power of user m on a kth upstream subchannels, represent the thermal noise that user n is subject on a kth upstream subchannels, represent the fading coefficients of user n on a kth upstream subchannels, at this symbol " || " for plural number asks modulo operation symbol, represent that user n loads on a kth upstream subchannels power required for individual bit, represent the power of Virtual User r on a kth upstream subchannels;

6.-5, from the weighted power of user n on all upstream subchannels, the minimum weighted power of value is found out;

6. the message transmission rate R that whether can cause user n after user n loads a bit on the upstream subchannels of the minimum weighted power of this value-6, is judged ⁿexceed the target data transmission rate R of user n ^{n, target}if meeting, then perform step 6.-9, otherwise, continue to perform step 6.-7, wherein, f _srepresent symbol rate;

6.-7, judge whether the weighted power of user n on the upstream subchannels of the minimum weighted power of this value is infinitely great, if so, then makes then return step and 6.-2 continue to perform, otherwise, continue to perform step 6.-8, wherein, in "=" be assignment;

6.-8, judge whether to cause after user n loads a bit on the upstream subchannels of the minimum weighted power of this value the power of user n on the upstream subchannels of the minimum weighted power of this value to exceed the maximum power restriction of user n on the upstream subchannels of the minimum weighted power of this value, if meeting, then the weighted power of user n on the upstream subchannels of the minimum weighted power of this value is set to infinity, then step 6.-5 continuation execution are returned, otherwise, user n loads a bit on the upstream subchannels of the minimum weighted power of this value, then the weighted power of user n on the upstream subchannels of the minimum weighted power of this value is upgraded, then step 6.-5 continuation execution are returned,

6.-9, judge the power of user n on all upstream subchannels and whether can exceed the up gross power restriction of user n, if meeting, then make then step 6.-2 continuation execution are returned, otherwise, order then step 6.-2 continuation execution are returned, wherein, with in "=" be assignment;

7. repeatedly successively perform step 4., step 5. with step 6., until convergence determines that weak user organizes bit number that each weak user in U1 loads on each upstream subchannels and strong user and organizes the bit number that each strong user in U2 loads on each upstream subchannels, that is: weak user organize in U1 each weak user before and after the difference of total bit number of two sub-distribution when being less than 0.5% of total bit number of primary distribution above, stop the above-mentioned strong user bit based on power weightings organized in U2 to load and weak user organizes RA-LC bit loading procedure in U1.

The described strong user Signal to Interference plus Noise Ratio sum of any one strong user on all upstream subchannels organized in U2 organizes the Signal to Interference plus Noise Ratio sum of each weak user on all upstream subchannels in U1 higher than described weak user.

Computational process for the Signal to Interference plus Noise Ratio of user n on a kth upstream subchannels is: be designated as by the Signal to Interference plus Noise Ratio of user n on a kth upstream subchannels ${\mathrm{SINR}}_k^n=\frac{1}{\mathrm{\Γ}}\frac{{\left|h_k^{n,n}\right|}^2{s}_k^n}{\underset{m\≠n}{\mathrm{\Σ}}{\left|h_k^{n,m}\right|}^2{s}_k^m+{\mathrm{\σ}}_k^n},$ Wherein, Γ is signal to noise ratio difference, represent the fading coefficients of user n on a kth upstream subchannels, represent the power of user n on a kth upstream subchannels, represent that user m is to the cross-talk coefficient of user n on a kth upstream subchannels, represent the power of user m on a kth upstream subchannels, represent the thermal noise that user n is subject on a kth upstream subchannels, at this symbol " || " for plural number asks modulo operation symbol, n=1,2 ..., N, m=1,2 ..., N, k=1,2 ..., K, N represent number of users, the number of the upstream subchannels in K representative digit subscribers feeder up-link.

Compared with prior art, the invention has the advantages that:

1) the inventive method adopts the bit loading method based on power weightings to carry out in the detailed process of bit distribution to all upstream subchannels, by utilizing strong user to the corresponding relation of the power division of the interference effect of weak user and weak user, the power increment of the power of weak user and strong user is weighted on average by strong user, then all choosing the minimum upstream subchannels of weighted power carries out bit loading at every turn, more bit can be loaded on high frequency subchannel like this, thus can effectively suppress strong user to the cross-talk of weak user.

2) the inventive method adopts the bit loading method based on power weightings to carry out in the detailed process of bit distribution to all upstream subchannels, by upgrading the weighted factor of strong user, can guarantee that strong user reaches the target data transmission rate of self, maximize the message transmission rate of weak user simultaneously.

Accompanying drawing explanation

Fig. 1 is that in the inventive method, weak user organizes modulator-demodulator corresponding to any one weak user in U1 and adopts RA-LC bit loading method all upstream subchannels to be carried out to the FB(flow block) of bit distribution;

Fig. 2 is that in the inventive method, strong user organizes modulator-demodulator corresponding to any one strong user in U2 and adopts the bit loading method based on power weightings all upstream subchannels to be carried out to the FB(flow block) of bit distribution;

Fig. 3 is the schematic diagram of two users VDSL2 uplink system;

Fig. 4 be the inventive method with the message transmission rate of existing RA-LC bit loading algorithm, optimum spectral equalization (OSB) algorithm, the preferential spectrum management of distributed frequency bands (DBPSM) algorithm compare schematic diagram.

Embodiment

Below in conjunction with accompanying drawing embodiment, the present invention is described in further detail.

Bit loading method in a kind of Digital Subscriber Line up-link that the present invention proposes, its processing procedure is: all users are divided into weak user according to the channel status of each user on all upstream subchannels and organize U1 and strong user organizes U2 two groups; Weak user organizes modulator-demodulator corresponding to each weak user in U1 and adopts RA-LC bit loading method to carry out bit distribution to all upstream subchannels; Strong user organizes modulator-demodulator corresponding to each strong user in U2 and adopts the bit loading method based on power weightings to carry out bit distribution to all upstream subchannels, and the main process of distribution is: first, calculates the weighted factor of strong user; Then obtain the weighted power of strong user on each upstream subchannels, find out the weighted power that value is minimum; Then, according to the target data transmission rate of strong user, the weighted power of strong user on the upstream subchannels being worth minimum weighted power, the maximum power of strong user on the upstream subchannels being worth minimum weighted power limits and the up gross power of strong user limits, and carries out bit distribution and upgrade weighted factor; Finally, according to the difference of the up gross power restriction of strong user and the maximum weighted Summing Factor minimum weight factor of strong user, end bit assigning process is determined whether.

Bit loading method of the present invention specifically comprises the following steps:

1. digital subscriber line system adopts Discrete Multitone modulation (DMT, DiscreteMulti-tone), whole channel is divided into multiple independently subchannel, again uplink and downlink division is carried out to all subchannels, to the subchannel of uplink information transmission be used for as upstream subchannels, will the subchannel of downlink information transmission be used for as downlink sub-channels.

2. digital subscriber line system is divided into Digital Subscriber Line up-link and Digital Subscriber Line downlink system according to uplink and downlink transmission direction.Spectrum Management Center (SMC) in Digital Subscriber Line up-link is according to the channel status of each user on all upstream subchannels, all users are divided into weak user and organize U1 and strong user organizes U2 two groups, the strong user channel status of any one strong user on whole up channel organized in U2 is better than weak user and organizes the channel status of each weak user on whole up channel in U1, and then user grouping information is sent to each modulator-demodulator in Digital Subscriber Line up-link by Spectrum Management Center.

At this, the strong user channel status of any one strong user on whole up channel organized in U2 is better than weak user and organizes the channel status of each weak user on whole up channel in U1, is: the strong user's Signal to Interference plus Noise Ratio sum of any one strong user on all upstream subchannels organized in U2 organizes the Signal to Interference plus Noise Ratio sum of each weak user on all upstream subchannels in U1 higher than weak user.At this, the Signal to Interference plus Noise Ratio of user n on a kth upstream subchannels is designated as ${\mathrm{SINR}}_k^n=\frac{1}{\mathrm{\Γ}}\frac{{\left|h_k^{n,n}\right|}^2{s}_k^n}{\underset{m\≠n}{\mathrm{\Σ}}{\left|h_k^{n,m}\right|}^2{s}_k^m+{\mathrm{\σ}}_k^n},$ Wherein, Γ is signal to noise ratio difference, represent the fading coefficients of user n on a kth upstream subchannels, represent the power of user n on a kth upstream subchannels, represent that user m is to the cross-talk coefficient of user n on a kth upstream subchannels, represent the power of user m on a kth upstream subchannels, represent the thermal noise that user n is subject on a kth upstream subchannels, at this symbol " || " for plural number asks modulo operation symbol, n=1,2,, N, m=1,2 ..., N, k=1,2 ... K, N represent the logarithm of the twisted-pair feeder in a cable, namely represent number of users, the number of the upstream subchannels in K representative digit subscribers feeder up-link, when 8.625KHz is got at upstream subchannels interval, the number K of total upstream subchannels is 1954.

3., in Digital Subscriber Line up-link, a spectrum management model is built:

$\underset{s^1,...,s^n,...,s^N}{\max }\underset{n\∈U1}{\mathrm{\Σ}}{R}^n$

R ⁿ≥R ^n,target,n∈U2

$0\≤s_k^n\≤s_k^{n,\mathrm{mask}},n=1,...,N,k=1,...,K,$ This spectrum management model description is reach given target strong user

$\underset{k=1}{\overset{K}{\mathrm{\Σ}}}{s}_k^n\≤P^{n,\mathrm{tot}},n=1,...,N$

When data transmission rates demands, maximize the message transmission rate of weak user, wherein, N represents number of users, the number of the upstream subchannels in K representative digit subscribers feeder up-link, max () for getting max function, S ¹represent the power allocation vector of user 1 on all upstream subchannels, S ⁿrepresent the power allocation vector of user n on all upstream subchannels, s ⁿrepresent the power allocation vector of user N on all upstream subchannels, represent the power of user n on the 1st upstream subchannels, represent the power of user n on a kth upstream subchannels, represent the power of user n on K upstream subchannels, U1 represents weak user's group, R ⁿrepresent the message transmission rate of user n, R ^{n, target}represent the target data transmission rate of user n, U2 represents strong user's group, represent the maximum power restriction of user n on a kth upstream subchannels, p ^{n, tot}represent the up gross power restriction of user n.

At this, the target data transmission rate that strong user organizes each the strong user in U2 is given, and the message transmission rate that weak user organizes each the weak user in U1 is not given.

4., in Digital Subscriber Line up-link, weak user organizes modulator-demodulator corresponding to each weak user in U1 and adopts RA-LC bit loading method to carry out bit distribution to all upstream subchannels.

In this particular embodiment, as shown in Figure 1, step 4. in weak user organize modulator-demodulator corresponding to any one weak user in U1 and adopt RA-LC bit loading method to the detailed process that all upstream subchannels carry out bit distribution to be:

-1 4., suppose that this weak user is user n, and obtained the power increment of user n on each upstream subchannels, suppose that user n is loaded with on a kth upstream subchannels individual bit, then load on a kth upstream subchannels by user n power increment required for individual bit is designated as wherein, n=1,2 ..., N, N represent number of users, k=1,2 ..., the number of the upstream subchannels in K, K representative digit subscribers feeder up-link, represent the bit number that user n loads on a kth upstream subchannels, order initial value be 0, represent that user n loads on a kth upstream subchannels power required for individual bit, $p_{\left(b_k^n\right)}=\mathrm{\Γ}(2^{b_k^n}-1)\frac{\underset{m\≠n}{\mathrm{\Σ}}{\left|h_k^{n,m}\right|}^2{s}_k^n+{\mathrm{\σ}}_k^n}{{\left|h_k^{n,n}\right|}^2},$ Γ is signal to noise ratio difference, represent that user m is to the cross-talk coefficient of user n on a kth upstream subchannels, represent the power of user m on a kth upstream subchannels, represent the thermal noise that user n is subject on a kth upstream subchannels, represent the fading coefficients of user n on a kth upstream subchannels, at this symbol " || " for plural number asks modulo operation symbol, represent that user n loads on a kth upstream subchannels power required for individual bit.

4.-2, from the power increment of user n on all upstream subchannels, the minimum power increment of value is found out.

4.-3, judge whether user n can cause the gross power of user n to exceed the up gross power restriction of user n load a bit on the upstream subchannels of the minimum power increment of this value after, if meeting, the modulator-demodulator that then user n is corresponding terminates the bit allocation procedures of all upstream subchannels, otherwise, continue to perform step 4.-4.

4.-4, judge whether to cause after user n loads a bit on the upstream subchannels of the minimum power increment of this value the power of user n on the upstream subchannels of the minimum power increment of this value to exceed the maximum power restriction of user n on the upstream subchannels of the minimum power increment of this value, if meeting, then the power increment of user n on the upstream subchannels of the minimum power increment of this value is set to infinity, then step 4.-2 continuation execution are returned, otherwise, user n loads a bit on the upstream subchannels of the minimum power increment of this value, then the power increment of user n on the upstream subchannels of the minimum power increment of this value is upgraded, then step 4.-2 continuation execution are returned.

5. fabricate a Virtual User r to represent all weak users, the power of Virtual User r on each upstream subchannels is determined by all power of weak user on each upstream subchannels, is designated as by the power of Virtual User r on a kth upstream subchannels wherein, k=1 ..., the number of the upstream subchannels in K, K representative digit subscribers feeder up-link, n=1,2 ..., N, N represent number of users, represent the power of user n on a kth upstream subchannels, d represents that weak user organizes the number of the weak user in U1.

The power allocation case of Virtual User r on each upstream subchannels sends to strong user to organize each strong user in U2 by the Spectrum Management Center 6. in Digital Subscriber Line up-link, and the modulator-demodulator that strong user organizes each strong user in U2 corresponding adopts the bit loading method based on power weightings to carry out bit distribution to all upstream subchannels.

In this particular embodiment, as shown in Figure 2, step 6. in strong user organize modulator-demodulator corresponding to any one strong user in U2 and adopt the bit loading method based on power weightings to the detailed process that all upstream subchannels carry out bit distribution to be:

-1 6., suppose that this strong user is user n, order represent the minimum weight factor of user n, and initial value be 0, order represent the maximum weighted factor of user n, and initial value be 1, make the power of user n on a kth upstream subchannels initial value be 0, k=1,2 ..., K.

6.-2, basis with calculate the weighted factor of user n, be designated as w ⁿ,

6.-3, judge $|\underset{k}{\mathrm{\Σ}}{s}_k^n-P^{n,\mathrm{tot}}|>P^{n.\mathrm{tot}}\×0.005$ With $w_{\max }^n-w_{\min }^n>0.0001$ Whether set up simultaneously, if set up, then perform step 6.-4, otherwise the modulator-demodulator that user n is corresponding terminates the bit allocation procedures of all upstream subchannels, wherein, P ^{n, tot}representing the up gross power restriction of user n, is the symbol that takes absolute value at this symbol " || ".

-4 6., obtain the weighted power of user n on each upstream subchannels, the weighted power of user n on kth sub-channels is designated as ${\mathrm{eff}}_k^n=w^n{\mathrm{\Δp}}_{(b_k^n+1)}+(1-w^n)s_k^r,$ Wherein, represent that user n loads the on a kth upstream subchannels power increment required for individual bit, n=1,2 ..., N, N represent number of users, k=1,2 ..., the number of the upstream subchannels in K, K representative digit subscribers feeder up-link, represent the bit number that user n loads on a kth upstream subchannels, order initial value be 0, represent that user n loads on a kth upstream subchannels power required for individual bit, $p_{\left(b_k^n\right)}=\mathrm{\Γ}(2^{b_k^n}-1)\frac{\underset{m\≠n}{\mathrm{\Σ}}{\left|h_k^{n,m}\right|}^2{s}_k^n+{\mathrm{\σ}}_k^n}{{\left|h_k^{n,n}\right|}^2},$ Γ is signal to noise ratio difference, represent that user m is to the cross-talk coefficient of user n on a kth upstream subchannels, represent the power of user m on a kth upstream subchannels, represent the thermal noise that user n is subject on a kth upstream subchannels, represent the fading coefficients of user n on a kth upstream subchannels, at this symbol " || " for plural number asks modulo operation symbol, represent that user n loads on a kth upstream subchannels power required for individual bit, represent that user n loads the on a kth upstream subchannels power increment required for individual bit, represent the power of Virtual User r on a kth upstream subchannels.

6.-5, from the weighted power of user n on all upstream subchannels, the minimum weighted power of value is found out.

6. the message transmission rate R that whether can cause user n after user n loads a bit on the upstream subchannels of the minimum weighted power of this value-6, is judged ⁿexceed the target data transmission rate R of user n ^{n, target}if, meeting, then the modulator-demodulator that user n is corresponding terminates the bit allocation procedures of this all upstream subchannels, then performs step 6.-9, otherwise, continue to perform step 6.-7, wherein, f _srepresent symbol rate.

6.-7, judge whether the weighted power of user n on the upstream subchannels of the minimum weighted power of this value is infinitely great, and if so, then the modulator-demodulator that user n is corresponding terminates the bit allocation procedures of this all upstream subchannels, and makes then return step and 6.-2 continue to perform, otherwise, continue to perform step 6.-8, wherein, in "=" be assignment.

6.-8, judge whether to cause after user n loads a bit on the upstream subchannels of the minimum weighted power of this value the power of user n on the upstream subchannels of the minimum weighted power of this value to exceed the maximum power restriction of user n on the upstream subchannels of the minimum weighted power of this value, if meeting, then the weighted power of user n on the upstream subchannels of the minimum weighted power of this value is set to infinity, then step 6.-5 continuation execution are returned, otherwise, user n loads a bit on the upstream subchannels of the minimum weighted power of this value, then the weighted power of user n on the upstream subchannels of the minimum weighted power of this value is upgraded, then step 6.-5 continuation execution are returned.

6.-9, judge the power of user n on all upstream subchannels and whether can exceed the up gross power restriction of user n, if meeting, then make then step 6.-2 continuation execution are returned, otherwise, order then step 6.-2 continuation execution are returned, wherein, with in "=" be assignment.

7. repeatedly step is performed 4. successively, step 5. with step 6., until convergence determines that weak user organizes bit number that each weak user in U1 loads on each upstream subchannels and strong user and organizes the bit number that each strong user in U2 loads on each upstream subchannels, that is: weak user organize in U1 each weak user before and after the difference of total bit number of two sub-distribution when being less than 0.5% of total bit number of primary distribution above, stop above-mentioned strong user to organize the bit based on power weightings in U2 to load and weak user organizes RA-LC bit loading procedure in U1, the bit number that each user loads on each upstream subchannels so just determines, also a kind of bit allocation scheme is just defined.

The data encoding transmitted needed for each user is become the binary bits number that can transmit by the transmitter 8. in Digital Subscriber Line up-link, the bit number that modulator-demodulator corresponding to each user loads on each upstream subchannels according to this user carries out bit distribution to needing the binary bits number of transmission, is then sent by channel by the bit distributed.

Be below the validity by the objective explanation the inventive method of Computer Simulation and feasibility.

Simulated conditions: two users' distributed topology structure of emulation as shown in Figure 3, user 1 is apart from optical network unit 600m, user 2 apart from optical network unit 300m, because user 2 is nearer apart from optical network unit than user 1, so user 2 couples of users 1 disturb comparatively user's 1 pair of user 2 serious interference.Adopt FranceTelecome(France Telecom portion) measured data that provides, upstream band divides and adopts 3.75MHz ~ 5.2MHz, 8.5MHz ~ 12MHz and 18.1MHz ~ 30MHz.Noise power spectrum elects-140dB as, sub-carrier frequency spacing is 8.625KHz, and symbol rate is 7KHz, and signal to noise ratio difference is 19.5236, the maximum power restriction of each user on each upstream subchannels is-56.5dB, and the up gross power restriction of all users is all 11dB.

Fig. 4 gives under above-mentioned simulated conditions, the message transmission rate of the inventive method and existing RA-LC bit loading algorithm, optimum spectral equalization (OSB) algorithm, the preferential spectrum management of distributed frequency bands (DBPSM) algorithm compare schematic diagram.As can be seen from Figure 4 the message transmission rate of the message transmission rate of the inventive method closely OSB algorithm, be higher than the message transmission rate of RA-LC bit loading algorithm and DBPSM algorithm, wherein the message transmission rate of OSB algorithm can reach maximum in theory, but OSB algorithm computation complexity is too high, be difficult to practical application.

Claims (3)

1. the bit loading method in Digital Subscriber Line up-link, is characterized in that its processing procedure is: all users are divided into weak user according to the channel status of each user on all upstream subchannels and organize U1 and strong user organizes U2 two groups; Weak user organizes modulator-demodulator corresponding to each weak user in U1 and adopts Rate-adaptiveLevin-Campello bit loading method to carry out bit distribution to all upstream subchannels; Strong user organizes modulator-demodulator corresponding to each strong user in U2 and adopts the bit loading method based on power weightings to carry out bit distribution to all upstream subchannels, and the main process of distribution is: first, calculates the weighted factor of strong user; Then obtain the weighted power of strong user on each upstream subchannels, find out the weighted power that value is minimum; Then, according to the target data transmission rate of strong user, the weighted power of strong user on the upstream subchannels being worth minimum weighted power, the maximum power of strong user on the upstream subchannels being worth minimum weighted power limits and the up gross power of strong user limits, and carries out bit distribution and upgrade weighted factor; Finally, according to the difference of the up gross power restriction of strong user and the maximum weighted Summing Factor minimum weight factor of strong user, end bit assigning process is determined whether;

This bit loading method specifically comprises the following steps:

1. digital subscriber line system adopts Discrete Multitone modulation, whole channel is divided into multiple independently subchannel, again uplink and downlink division is carried out to all subchannels, will the subchannel of uplink information transmission be used for as upstream subchannels, will the subchannel of downlink information transmission be used for as downlink sub-channels;

2. the Spectrum Management Center in Digital Subscriber Line up-link is according to the channel status of each user on all upstream subchannels, all users are divided into weak user and organize U1 and strong user organizes U2 two groups, then user grouping information is sent to each modulator-demodulator in Digital Subscriber Line up-link by Spectrum Management Center;

3., in Digital Subscriber Line up-link, a spectrum management model is built: $\underset{S^1,...,S^n,...,S^N}{max}\underset{n\∈U1}{\Σ}{R}^n$

R ⁿ≥R ^n,target,n∈U2

$0\≤s_k^n\≤s_k^{n,mask},n=1,...,N,k=1,...,K,$ This spectrum management model description for strong user reach to $\underset{k=1}{\overset{K}{\Σ}}{s}_k^n\≤P^{n,tot},n=1,...,N$

When fixed target data transmission rate request, maximize the message transmission rate of weak user, wherein, N represents number of users, the number of the upstream subchannels in K representative digit subscribers feeder up-link, max () for getting max function, S ¹represent the power allocation vector of user 1 on all upstream subchannels, S ⁿrepresent the power allocation vector of user n on all upstream subchannels, s ⁿrepresent the power allocation vector of user N on all upstream subchannels, represent the power of user n on the 1st upstream subchannels, represent the power of user n on a kth upstream subchannels, represent the power of user n on K upstream subchannels, U1 represents weak user's group, R ⁿrepresent the message transmission rate of user n, R ^{n, target}represent the target data transmission rate of user n, U2 represents strong user's group, represent the maximum power restriction of user n on a kth upstream subchannels, P ^{n, tot}represent the up gross power restriction of user n;

4., in Digital Subscriber Line up-link, weak user organizes modulator-demodulator corresponding to each weak user in U1 and adopts Rate-adaptiveLevin-Campello bit loading method to carry out bit distribution to all upstream subchannels;

At this, weak user organizes modulator-demodulator corresponding to any one weak user in U1 and adopts Rate-adaptiveLevin-Campello bit loading method to the detailed process that all upstream subchannels carry out bit distribution to be:

-1 4., suppose that this weak user is user n, and obtained the power increment of user n on each upstream subchannels, suppose that user n is loaded with on a kth upstream subchannels individual bit, then load on a kth upstream subchannels by user n power increment required for individual bit is designated as wherein, n=1,2 ..., N, N represent number of users, k=1,2 ..., the number of the upstream subchannels in K, K representative digit subscribers feeder up-link, represent the bit number that user n loads on a kth upstream subchannels, order initial value be 0, represent that user n loads on a kth upstream subchannels power required for individual bit, Γ is signal to noise ratio difference, represent that user m is to the cross-talk coefficient of user n on a kth upstream subchannels, represent the power of user m on a kth upstream subchannels, represent the thermal noise that user n is subject on a kth upstream subchannels, represent the fading coefficients of user n on a kth upstream subchannels, at this symbol " || " for plural number asks modulo operation symbol, represent that user n loads on a kth upstream subchannels power required for individual bit;

4.-2, from the power increment of user n on all upstream subchannels, the minimum power increment of value is found out;

4.-3, judge whether user n can cause the gross power of user n to exceed the up gross power restriction of user n load a bit on the upstream subchannels of the minimum power increment of this value after, if meeting, the modulator-demodulator that then user n is corresponding terminates the bit allocation procedures of all upstream subchannels, otherwise, continue to perform step 4.-4;

4.-4, judge whether to cause after user n loads a bit on the upstream subchannels of the minimum power increment of this value the power of user n on the upstream subchannels of the minimum power increment of this value to exceed the maximum power restriction of user n on the upstream subchannels of the minimum power increment of this value, if meeting, then the power increment of user n on the upstream subchannels of the minimum power increment of this value is set to infinity, then step 4.-2 continuation execution are returned, otherwise, user n loads a bit on the upstream subchannels of the minimum power increment of this value, then the power increment of user n on the upstream subchannels of the minimum power increment of this value is upgraded, then step 4.-2 continuation execution are returned,

5. fabricate a Virtual User r to represent all weak users, the power of Virtual User r on each upstream subchannels is determined by all power of weak user on each upstream subchannels, is designated as by the power of Virtual User r on a kth upstream subchannels wherein, k=1 ..., the number of the upstream subchannels in K, K representative digit subscribers feeder up-link, n=1,2 ..., N, N represent number of users, represent the power of user n on a kth upstream subchannels, d represents that weak user organizes the number of the weak user in U1;

The power allocation case of Virtual User r on each upstream subchannels sends to strong user to organize each strong user in U2 by the Spectrum Management Center 6. in Digital Subscriber Line up-link, and the modulator-demodulator that strong user organizes each strong user in U2 corresponding adopts the bit loading method based on power weightings to carry out bit distribution to all upstream subchannels;

At this, strong user organizes modulator-demodulator corresponding to any one strong user in U2 and adopts the bit loading method based on power weightings to the detailed process that all upstream subchannels carry out bit distribution to be:

-1 6., suppose that this strong user is user n, order represent the minimum weight factor of user n, and initial value be 0, order represent the maximum weighted factor of user n, and initial value be 1, make the power of user n on a kth upstream subchannels initial value be 0, k=1,2 ..., K;

6.-2, basis with calculate the weighted factor of user n, be designated as w ⁿ,

6.-3, judge with whether set up simultaneously, if set up, then perform step 6.-4, otherwise the modulator-demodulator that user n is corresponding terminates the bit allocation procedures of all upstream subchannels, wherein, P ^{n, tot}representing the up gross power restriction of user n, is the symbol that takes absolute value at this symbol " || ";

-4 6., obtain the weighted power of user n on each upstream subchannels, the weighted power of user n on kth sub-channels is designated as wherein, represent that user n loads the on a kth upstream subchannels power increment required for individual bit, ${\mathrm{\Δp}}_{(b_k^n+1)}=p_{(b_k^n+1)}-p_{\left(b_k^n\right)},n=1,2,...,N,$ N represents number of users, k=1,2 ..., the number of the upstream subchannels in K, K representative digit subscribers feeder up-link, represent the bit number that user n loads on a kth upstream subchannels, order initial value be 0, represent that user n loads on a kth upstream subchannels power required for individual bit, $p_{\left(b_k^n\right)}=\mathrm{\Γ}(2^{b_k^n}-1)\frac{\underset{m\≠n}{\Σ}|h_k^{n,m}{|}^2{s}_k^m+{\mathrm{\σ}}_k^n}{|h_k^{n,n}{|}^2},$ Γ is signal to noise ratio difference, represent that user m is to the cross-talk coefficient of user n on a kth upstream subchannels, represent the power of user m on a kth upstream subchannels, represent the thermal noise that user n is subject on a kth upstream subchannels, represent the fading coefficients of user n on a kth upstream subchannels, at this symbol " || " for plural number asks modulo operation symbol, represent that user n loads on a kth upstream subchannels power required for individual bit, represent the power of Virtual User r on a kth upstream subchannels;

6.-5, from the weighted power of user n on all upstream subchannels, the minimum weighted power of value is found out;

6. the message transmission rate R that whether can cause user n after user n loads a bit on the upstream subchannels of the minimum weighted power of this value-6, is judged ⁿexceed the target data transmission rate R of user n ^{n, target}if meeting, then perform step 6.-9, otherwise, continue to perform step 6.-7, wherein, f _srepresent symbol rate;

6.-7, judge whether the weighted power of user n on the upstream subchannels of the minimum weighted power of this value is infinitely great, if so, then makes then return step and 6.-2 continue to perform, otherwise, continue to perform step 6.-8, wherein, in "=" be assignment;

6.-8, judge whether to cause after user n loads a bit on the upstream subchannels of the minimum weighted power of this value the power of user n on the upstream subchannels of the minimum weighted power of this value to exceed the maximum power restriction of user n on the upstream subchannels of the minimum weighted power of this value, if meeting, then the weighted power of user n on the upstream subchannels of the minimum weighted power of this value is set to infinity, then step 6.-5 continuation execution are returned, otherwise, user n loads a bit on the upstream subchannels of the minimum weighted power of this value, then the weighted power of user n on the upstream subchannels of the minimum weighted power of this value is upgraded, then step 6.-5 continuation execution are returned,

6.-9, judge the power of user n on all upstream subchannels and whether can exceed the up gross power restriction of user n, if meeting, then make then step 6.-2 continuation execution are returned, otherwise, order then step 6.-2 continuation execution are returned, wherein, with in "=" be assignment;

7. repeatedly step is performed 4. successively, step 5. with step 6., until convergence determines that weak user organizes bit number that each weak user in U1 loads on each upstream subchannels and strong user and organizes the bit number that each strong user in U2 loads on each upstream subchannels, that is: weak user organize in U1 each weak user before and after the difference of total bit number of two sub-distribution when being less than 0.5% of total bit number of primary distribution above, stop above-mentioned strong user to organize the bit based on power weightings in U2 to load and weak user organizes Rate-adaptiveLevin-Campello bit loading procedure in U1.

2. the bit loading method in a kind of Digital Subscriber Line up-link according to claim 1, is characterized in that the described strong user Signal to Interference plus Noise Ratio sum of any one strong user on all upstream subchannels organized in U2 organizes the Signal to Interference plus Noise Ratio sum of each weak user on all upstream subchannels in U1 higher than described weak user.

3. the bit loading method in a kind of Digital Subscriber Line up-link according to claim 2, is characterized in that the computational process for the Signal to Interference plus Noise Ratio of user n on a kth upstream subchannels is: be designated as by the Signal to Interference plus Noise Ratio of user n on a kth upstream subchannels wherein, Γ is signal to noise ratio difference, represent the fading coefficients of user n on a kth upstream subchannels, represent the power of user n on a kth upstream subchannels, represent that user m is to the cross-talk coefficient of user n on a kth upstream subchannels, represent the power of user m on a kth upstream subchannels, represent the thermal noise that user n is subject on a kth upstream subchannels, at this symbol " || " for plural number asks modulo operation symbol, n=1,2 ..., N, m=1,2 ..., N, k=1,2 ..., K, N represent number of users, the number of the upstream subchannels in K representative digit subscribers feeder up-link.