KR100513340B1 - An adaptive bit rate allocation method using the estimation signal to noise ratio - Google Patents

Abstract

According to the present invention, adaptive bit allocation for each channel through signal-to-noise ratio estimation that adjusts the number of bits allocated to subcarriers (channels) of a DMT channel in real time according to a transmission state of a channel in a DMT (Discrete Multi-Tone) communication system The purpose is to provide a method.

The present invention provides a method for allocating the number of bits according to a reception state of each channel in a DMT communication system for transmitting and receiving data through a plurality of channels using a plurality of subcarriers, the power spectrum of the received signal of each channel. Obtaining an average value PM and a dispersion value PD of the density; Iteratively obtaining a signal-to-noise ratio measurement value (SNRme) by dividing the average value (PM) of the power spectrum density by a variance value (PD); Recursively obtaining the signal-to-noise ratio measurement value SNRme, and then obtaining an average value SNRav of the measurement value SNRme; Obtaining an error Err between the signal-to-noise ratio measurement value SNRme and the signal-to-noise ratio average value SNRav; Updating the signal-to-noise ratio average value SNRav using the error Err; Updating a preset signal-to-noise ratio variance value Dev using the error Err; Estimating a signal-to-noise ratio (SNRout) for a next received signal using the updated dispersion value Dev and the average value of the signal-to-noise ratio SNRav; And allocating the number of bits for the corresponding channel according to the estimated signal-to-noise ratio (SNRout).

According to the present invention, after the first link is formed between the transmitting and receiving end, even if the data rate that can be provided in each DMT channel changes due to a change in channel environment, it is possible to quickly adapt.

Description

Adaptive Bit Allocation Method for Each Channel by Estimating Signal-to-Noise Ratio {AN ADAPTIVE BIT RATE ALLOCATION METHOD USING THE ESTIMATION SIGNAL TO NOISE RATIO}

The present invention relates to an adaptive bit allocation method for each channel through signal-to-noise ratio estimation in a DMT communication system. In particular, in a DMT (Discrete Multi-Tone) communication system, a subcarrier of a DMT channel according to a transmission state of a channel By adjusting the number of bits allocated to the channel in real time, the signal-to-noise ratio estimation can be quickly adapted even if the data rate provided by each DMT channel changes due to the change of channel environment after the first link formation between the transmitting and receiving ends. It relates to an adaptive bit allocation method for each channel.

In general, a communication system generates a difference in data rates that can be provided according to a link state between an external environment and a transmitting / receiving end. For example, a better channel environment enables faster data communication.

One of the conventional channel-specific bit allocation methods measures the power spectrum density to obtain an SNR, and compares it with the existing SNR to modify the bit and power allocation tables of the transmitting and receiving ends.

A conventional bit exchange method is shown in FIG.

1 is a flow chart of a conventional bit exchange process. Referring to FIG. 1, a process flow of a conventional bit exchange process is described. Referring to FIG. 1, a codeword number of a frame in which an error occurs from an FEC decoder during reception, a number of codewords for correcting an error, Received error information such as the number of codewords that cannot be corrected is judged, and if it is greater than or equal to the reference value, the bit exchange is processed, and if it is smaller, the existing bit table is maintained.

More specifically, in step S1, the error of the super frame received by the FEC decoder of the DMT receiving unit is checked, and the number of FEC code words in which an error occurs every super frame, and an FEC code correcting the error every super frame. The number of words and the number of FEC code words whose errors cannot be corrected every super frame are provided to the DMT control unit. In step S2, the receiving side DMT control unit compares the values of the error information received from the FEC decoder (this is simply called an error value) with a preset reference value, and if the error value is greater than or equal to the reference value, processes the bit exchange according to the present invention. If it is smaller than the reference value, the current bit table is kept as it is in step S7.

Here, the error information detected by the FEC decoder indicates the state of the transport channel. If the error value is equal to or larger than the reference value, a bit swap request message is sent to the counterpart in step S3. The bit exchange request message consists of a 1 byte header and an n byte message field according to the AOC message encoding procedure. The AOC message is transmitted in a sync byte of the super frame described with reference to FIG. 5. Subsequently, when the DMT control unit receives the bit exchange response message while checking whether the bit exchange response message is received from the counterpart while waiting, the DMT control unit exchanges the bit table in step S5.

One conventional signal-to-noise ratio (SNR) estimation method is shown in FIG.

FIG. 2 is a block diagram showing a conventional signal-to-noise ratio (SNR) estimation method. Referring to FIG. 2, the signal-to-noise ratio (SNR) estimation in the conventional virtual experiment circuit 220 will be described. In blocks 204, 206, 208, and 210, each carrier is used. Equation output is subtracted from the symbol decision value or subtracted from the existing training information of the received training information. If the carrier is transmitted with a constant power series, the SNR can be determined by the average value of the noise power series estimate. The low pass filter (LPF) 212,214, 216,218 can then be used to smooth the SNR estimate. At this time, a reliable estimate is selected from the SNR information obtained from the data information and the SNR information based on the training sequence, and used to determine the trellis decoder and convergence factor.

However, in the conventional signal-to-noise ratio (SNR) estimation and bit-switching method as described above, if the channel environment is changed due to the change of connected devices even after the first link is established, the bits allocated to each DMT channel can be provided. The transmission rate is also different. In this case, if a larger amount of data is transmitted than a data rate that can be provided at a particular moment, a transmission error occurs and eventually degrades the overall system performance.

For this reason, there is a need for a method capable of effectively and rapidly changing the transmission rate and the number of bits allocated to the DMT channel according to the change of the channel environment.

The present invention has been proposed to solve the above problems, and an object thereof is to provide a real-time number of bits allocated to subcarriers (channels) of a DMT channel according to a transmission state of a channel in a DMT (Discrete Multi-Tone) communication system. After the first link is formed between the transmitting and receiving end, it provides an adaptive bit allocation method for each channel by estimating the signal-to-noise ratio that can be quickly adapted even if the data rate available in each DMT channel changes due to the change of channel environment. It is.

In order to achieve the above object of the present invention, the channel-specific adaptive bit allocation method through the signal-to-noise ratio estimation of the present invention

In the DMT communication system for transmitting and receiving data in a plurality of channels using a plurality of subcarriers, in the method of allocating the number of bits according to the reception state of each channel,

Obtaining an average value (PM) and a dispersion value (PD) of the power spectrum density for the received signal of each channel;

Iteratively obtaining a signal-to-noise ratio measurement value (SNRme) by dividing the average value (PM) of the power spectrum density by a variance value (PD);

Recursively obtaining the signal-to-noise ratio measurement value SNRme, and then obtaining an average value SNRav of the measurement value SNRme;

Obtaining an error Err between the signal-to-noise ratio measurement value SNRme and the signal-to-noise ratio average value SNRav;

Updating the signal-to-noise ratio average value SNRav using the error Err;

Updating a preset signal-to-noise ratio variance value Dev using the error Err;

Estimating a signal-to-noise ratio (SNRout) for a next received signal using the updated dispersion value Dev and the average value of the signal-to-noise ratio SNRav; And

Allocating the number of bits for the corresponding channel according to the estimated signal-to-noise ratio (SNRout)

Characterized in having a.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

3 is a schematic diagram of a DMT communication system for carrying out the present invention.

Referring to FIG. 3, the DMT communication system for implementing the present invention obtains an average value PM and a dispersion value PD of power spectrum density for a received signal of each channel, and calculates an average value PM of the power spectrum density. The signal-to-noise ratio measurement value (SNRme) is repeatedly obtained by dividing by the variance value (PD), and the signal-to-noise ratio measurement value (SNRme) is repeatedly obtained, and then the signal-to-noise ratio estimator (SNRav) is obtained. 310, a signal-to-noise ratio error calculator 320 for obtaining an error Err between the signal-to-noise ratio measurement value SNRme obtained from the signal-to-noise ratio estimator 310 and the signal-to-noise ratio average value SNRav, and the signal-to-noise ratio error The signal-to-noise ratio average value SNRav and the signal-to-noise ratio variance value Dev are updated using the error Err calculated by the calculator 320, and the updated dispersion value Dev and the average value of the signal-to-noise ratio SNRav are updated. )of And a channel mapper 330 for estimating a signal-to-noise ratio (SNRout) for the next received signal and allocating the number of bits for the corresponding channel according to the estimated signal-to-noise ratio (SNRout).

4 is a flowchart showing a method for allocating bits according to the present invention, and FIG. 5 is an explanatory diagram illustrating a method for estimating a signal-to-noise ratio according to the present invention.

Hereinafter, the operation and effects of the present invention will be described in detail with reference to the accompanying drawings.

2 to 5, the adaptive bit allocation method for each channel through the signal-to-noise ratio estimation according to the present invention, in the DMT communication system for transmitting and receiving data in a plurality of channels using a plurality of subcarriers, The number of bits is allocated according to the reception state.

Referring to FIG. 5, first, an average value PM and a variance value PD of power spectrum densities of received signals of respective channels are obtained (S51), and then the average value PM of the power spectrum density is calculated as a dispersion value PD. The signal-to-noise ratio measurement (SNRme) is repeatedly obtained (S52), that is, the cross correlation function between the sequence used for synchronization and the previously determined sequence among the signals received through the channel. After the power spectrum density is obtained, the average value and the variance of the power spectrum density are obtained using the fact that the sequence is repeated for a plurality of periods. This will be described with reference to FIG. 5.

Referring to FIG. 5, in the multicarrier signal-to-noise ratio (SNR) measuring method according to the present invention, after obtaining power spectrum density (PSD) of all subcarriers through a preamble symbol of each channel, Using this, the signal-to-noise ratio is estimated by observing the power change of a specific multicarrier for a short time, wherein the average of the power density is estimated as the signal strength and the variance as the noise strength, respectively. The signal-to-noise ratio (SNR) is obtained from the ratio of signal strength to noise strength. On the other hand, since it is necessary to estimate the strength of the signal and the noise with a small sample for a short time, the estimated value moves irregularly with time as shown in FIG.

On the other hand, the SNR determined as described above is an estimate of the channel information at a specific time, it is difficult to include the trend of the channel change, it is likely that an error occurs in the packet immediately received when modifying the bit table using this. Therefore, the updated signal-to-noise ratio (SNR) information is used in consideration of the change in the past channel characteristics through the process described below.

Then, the signal-to-noise ratio measurement value SNRme is repeatedly obtained, and then the average value SNRav of the measurement value SNRme is obtained (S53), and then the signal-to-noise ratio measurement value SNRme and the signal-to-noise ratio average value SNRav are obtained. The error Err is obtained as shown in Equation 1 below (S54).

[Equation 1]

Err = SNRme-SNRav

Here, SNRme is an SNR measurement value, SNRav is an average evaluation value of SNR, and Err means "actual measurement value-current SNR measurement value.

Next, the signal-to-noise ratio average value SNRav is updated using the error Err (S55), and the signal-to-noise ratio average value SNRav update is described in more detail as shown in Equation 2 below. The signal-to-noise ratio average value SNRav is updated by adding the value obtained by multiplying the error Err by a predetermined average coefficient g to the signal-to-noise ratio average value SNRav.

[Equation 2]

SNRav = SNRav + g × Err

Where g is the coefficient of the mean.

Then, the signal-to-noise ratio variance value Dev, which is previously set using the error Err, is updated as shown in Equation 3 (S56), that is, the signal-to-noise ratio variance value Dev is updated to the error Err. The signal-to-noise ratio variance value Dev is obtained by subtracting the predetermined signal-to-noise ratio variance value Dev from the absolute value of the multiplier and multiplying the predetermined deviation coefficient h by adding the multiplied value to the signal-to-noise variance value Dev. Update the.

[Equation 3]

Dev = Dev + h (| Err | -Dev)

Here, Dev is a variance value indicating a deviation of the mean, and h is a coefficient of deviation.

Then, using the updated dispersion value Dev and the average value SNRav of the signal-to-noise ratio, the signal-to-noise ratio SNRout for the next received signal is estimated as shown in Equation 4 below, i.e., the signal-to-noise ratio (SNRout) Estimation is obtained by subtracting the updated dispersion value Dev multiplied by a predetermined evaluation value β from the signal-to-noise ratio average value SNRav to obtain a signal-to-noise ratio SNRout for the next received signal of the channel. Estimate).

[Equation 4]

SNRout = SNRav-β × Dev

Here, SNRout is an SNR value to be used for bit loading, and β is a coefficient of an evaluation value.

Finally, the number of bits for the channel is allocated according to the estimated signal-to-noise ratio (SNRout) (S58). If the signal-to-noise ratio (SNRout) is a large value, a large number of bits is allocated, whereas When the noise ratio (SNRout) is a small value, a small number of bits is allocated. Specifically, the allocated number of bits may be different depending on the characteristics of the system to be applied, and when applied to a specific system, the value of the signal-to-noise ratio (SNRout) may be different. Accordingly, the number of bits allocated can be set in advance.

As described above, in the case of a subcarrier having a large change in the signal-to-noise ratio (SNR), since the deviation is large, the characteristic is considered when determining the final signal-to-noise ratio (SNR) value and is determined to be lower than the current estimate. Even if it suddenly has a low signal-to-noise ratio (SNR), stable communication is possible.

According to the present invention as described above, in the communication system of the DMT (Discrete Multi-Tone) system, by adjusting the number of bits allocated to the sub-carrier (channel) of the DMT channel according to the transmission state of the channel in real time, After the first link formation, even if the data rate that can be provided in each DMT channel changes due to a change in the channel environment, it is possible to quickly adapt.

In addition, the channel change can be accumulated and reflected, and the change rate of the channel characteristic is applied when estimating the SNR, thereby enabling relatively stable high-speed data transmission even when using a subcarrier having a large change in the characteristic.

Since the above description is only a description of specific embodiments of the present invention, the present invention is not limited to these specific embodiments, and various changes and modifications of the configuration are possible from the above-described specific embodiments of the present invention. It will be apparent to those skilled in the art to which the present invention pertains.

1 is a flowchart of processing a conventional bit exchange.

2 is a block diagram showing a conventional signal-to-noise ratio (SNR) estimation method.

3 is a schematic diagram of a DMT communication system for carrying out the present invention;

4 is a flowchart showing a method for allocating an adaptive bit for each channel according to the present invention.

5 is an explanatory diagram illustrating a signal-to-noise ratio estimation method according to the present invention.

Explanation of symbols on main parts of drawing

310: signal to noise ratio estimator

320: Signal to noise ratio error calculator

330: Channel Mapper

Claims (4)

In the DMT communication system for transmitting and receiving data in a plurality of channels using a plurality of subcarriers, in the method of allocating the number of bits according to the reception state of each channel, Obtaining an average value PM and a variance value PD of power spectrum densities of the received signals of each channel (S51); Obtaining a signal-to-noise ratio measurement value SNRme by dividing the average value PM of the power spectrum density by the variance value PD (S52); Resolving the signal-to-noise ratio measurement value SNRme repeatedly and then obtaining an average value SNRav of the measurement value SNRme (S53); Obtaining an error Err between the signal-to-noise ratio measurement value SNRme and the signal-to-noise ratio average value SNRav (S54); Updating the signal-to-noise ratio average value SNRav using the error Err (S55); Updating the preset signal-to-noise ratio variance value Dev using the error Err (S56); Estimating a signal-to-noise ratio (SNRout) for a next received signal using the updated dispersion value Dev and the average value of the signal-to-noise ratio SNRav (S57); And Allocating the number of bits for the corresponding channel according to the estimated signal-to-noise ratio (SNRout) (S58) Adaptive bit allocation method for each channel through signal-to-noise ratio estimation characterized in that it comprises. The method of claim 1, wherein the signal-to-noise ratio average value SNRav updating step S55 is performed. The signal-to-noise ratio average value (SNRav) is updated by adding the value obtained by multiplying the error (Err) by a predetermined average coefficient (g) to the signal-to-noise ratio average value (SNRav). Bit allocation method. The method of claim 1, wherein the signal-to-noise ratio variance value Dev update step S56 is performed. The signal-to-noise ratio variance value Dev is multiplied by a value obtained by subtracting a predetermined signal-to-noise ratio variance value Dev from the absolute value of the error Err, and added to the signal-to-noise ratio variance value Dev. Adaptive bit allocation method for each channel through signal-to-noise ratio estimation, characterized by updating the variance value Dev. The method of claim 1, wherein the signal-to-noise ratio (SNRout) estimating step (S57) Estimating the signal-to-noise ratio (SNRout) for the next received signal of the corresponding channel by subtracting the value of the updated dispersion value Dev multiplied by a predetermined evaluation value β from the signal-to-noise ratio average value (SNRav). Adaptive bit allocation method for each channel through signal-to-noise ratio estimation.