KR20050042717A - Method and apparatus of combined power control and error-correcting coding - Google Patents

Abstract

The present invention relates to a method and apparatus for combining power control and error correction code in a mobile communication system. The mobile communication terminal obtains information on an estimated channel power gain for a channel to which a code symbol coded by an error correction code is to be transmitted, and, if the channel power gain is equal to or greater than a predetermined power cutoff threshold, the code symbol for the code symbol. Allocate power determined according to channel power gain. On the other hand, if the channel power gain is not greater than or equal to the power cutoff threshold, power is cut off by allocating zero power to the code symbol. The present invention can maximize coding gain and power gain by combining power control and error correction code.

Description

METHOD AND APPARATUS OF COMBINED POWER CONTROL AND ERROR-CORRECTING CODING}

The present invention relates to a mobile communication system, and more particularly, to a method and apparatus for combining power control and error correction code.

The next generation mobile communication terminal, which aims to support high-speed data communication, requires a longer usage time than a conventional terminal mainly focused on voice calls. Therefore, various methods for minimizing the power consumption of the terminal and using the power more efficiently have been studied in order to increase the continuous usable time by charging the battery once.

Mobile communication systems such as synchronous CDMA (Code Division Multiple Access) 2000 and asynchronous Universal Mobile Telecommunication Service (UMTS) use channel encoding for reliable transmission of multimedia data. Channel coding is known to show very good information recovery performance in terms of bit error rate (BER) even at a low signal-to-noise ratio, and can be corrected even when an error of some bits occurs in a fading environment. In the next generation mobile communication systems, turbo coding (Turbo Coding) and low density parity check (LDPC) codes having performances approaching the Shannon limit known as the error correction capability are mainly considered as error correction codes.

In general, power control in a communication system serves to keep a received signal to noise ratio (SNR) constant for all code symbols, and an error correction code is received. It corrects errors occurring between symbols. These two technologies are considered separate and have been studied separately.

However, errors that occur within the limits of the ability of the error correction code to correct when the error correction code is used are corrected regardless of the received SNR, so the received SNR does not need to remain constant for all code symbols. Therefore, there is a need to consider a technique of reallocating and transmitting power between code symbols in terms of minimizing the probability of error after decoding.

Accordingly, an object of the present invention, which was devised to solve the problems of the prior art operating as described above, is to provide a method and apparatus for reallocating and transmitting power of code symbols according to channel conditions.

It is an object of the present invention to provide a method and apparatus for allocating transmit power of turbo and LDPC code symbols in terms of minimizing the probability of error after decoding.

In order to achieve the above object, an embodiment of the present invention provides a method of power control of a code symbol encoded by an error correction code in a mobile communication system.

Obtaining information about an estimated channel power gain for a channel to which a code symbol coded by an error correction code is to be transmitted;

Allocating power determined according to the channel power gain to the code symbol when the channel power gain is equal to or greater than a predetermined power cutoff threshold value;

And if the channel power gain is not equal to or greater than the power cutoff threshold, allocating zero power to the code symbol to cut off power.

Hereinafter, with reference to the accompanying drawings will be described in detail the operating principle of the preferred embodiment of the present invention. In the following description of the present invention, detailed descriptions of well-known functions or configurations will be omitted if it is determined that the detailed description of the present invention may unnecessarily obscure the subject matter of the present invention. Terms to be described later are terms defined in consideration of functions in the present invention, and may be changed according to intentions or customs of users or operators. Therefore, the definition should be made based on the contents throughout the specification.

The present invention described below cuts power for code symbols that will experience channel power gains less than a predetermined power cut threshold and allocates more power to the remaining code symbols that will experience channel power gains above the power cut threshold. This improves the overall reliability of the received signals at the receiving end.

1 illustrates a transmission and reception structure of a next generation mobile communication system according to an embodiment of the present invention.

Referring to FIG. 1, the information sources 10 generate and output an information sequence to be transmitted to a counterpart receiver. The channel encoder 12 outputs binary code symbols by encoding the input information sequence by a turbo code or an LDPC code according to a predetermined coding rate.

The code symbols are interleaved by the channel interleaver 14 according to a predetermined interleaving rule, and by the modulator 16, quadrature phase shift keying (QPSK), 8-ary PSK (8PSK), and 16-ary quadrature amplitude modulation (16QAM). In accordance with a modulation scheme such as 64 QAM (64-ary QAM), in-phase elements (I) and quadrature elements (Q) are converted into modulation symbols. A power reallocator 18 assigns the modulation symbols transmit power P (g) according to the channel power gain g of the corresponding channel 20. Although not shown, the channel power gain g is measured by the receiver's channel estimator and fed back to the transmitter.

The power controlled modulation symbols are delivered to the receiving end via channel 20. The symbols received at the receiver have a received power expressed by g'P (g) according to the actual channel gain g 'of the channel 20. In an ideal case, the actual channel gain should be exactly the same as the estimated channel gain, but in reality it may be slightly different due to the change of the channel and the estimation error, so that g 'is labeled to be distinguished from the estimated gain g.

The demodulator 30 demodulates the received modulation symbols according to a demodulation scheme corresponding to the modulation scheme used by the modulator 16, and a channel deinterleaver 32 is outputted from the demodulator 30. The code symbols are returned in their original order according to the deinterleaving rule corresponding to the interleaving rule used in the channel interleaver 14. The channel decoder 34 decodes the deinterleaved code symbols from the channel deinterleaver 32 according to an error correcting code used in the channel encoder 12, that is, a turbo code or an LDPC code. )

As is known, a turbo code or LDPC code may include code symbols of one systematic part representing an input information sequence according to a code rate and at least one parity portion for error correction of the systematic part ( Create code symbols of Parity Part). For example, when a code rate of 1/3 is used, one systematic code symbol and two parity code symbols are generated for one bit.

In the above-described transmission / reception structure, in order for the power gain g of the channels to be independent and have the same distribution (Independent, Identically Distributed, iid), the channel interleaver 14 having an interleaving length that is much larger than the coherence time of the channel. And channel deinterleaver 32 are used, and coded binary code symbols are modulated and transmitted, respectively. In this case, in case of Rayleigh fading, the channel power gain g has a probability density function as shown in Equation 1 below.

here, = E [g] and E [] means the mean function.

2 is a flowchart illustrating an operation of allocating power to a sign symbol according to an embodiment of the present invention.

Referring to FIG. 2, in step 40, information about an estimated channel gain g for a channel that a code symbol will undergo is fed back from the receiving end to the transmitting end. In step 42 the channel gain g is compared with a predetermined power off threshold g ₀ . If the channel gain g is greater than or equal to the power cutoff threshold g ₀ , in step 44, the transmitter allocates and transmits power P (g) to the corresponding code symbol. Otherwise, in step 46, the transmitting end does not allocate transmit power to the corresponding code symbol. That is, P (g) = 0. In step 48 the sign symbol is transmitted with the assigned power P (g).

The receiver performs turbo or LDPC decoding on the received code symbols. At this time, if there is no signal received in the time interval to expect the reception of the code symbol, the receiving end treats the corresponding code symbol as an (erasure) and recovers it through a turbo or LDPC decoding process. In this case, turbo or LDPC decoding performs general soft-decision based decoding. In particular, in case of LDPC decoding, hard-decision based decoding considering lost bits is also possible.

In a preferred embodiment of the present invention, when the channel power gain of the channel to be experienced by the sign symbol is g, the power P (g) allocated to the sign symbol in the process 44 is determined as shown in Equation 2 below.

g ₀ , P ₁ , and P ₂ are set to optimal values in terms of minimizing the probability of error after decoding the symbol symbols that have undergone such a channel, like stochastic modeling of the channel. At this time, the average value of the transmission powers is kept constant. For example, in the case of the Rayleigh fading channel, the average transmit power becomes as shown in Equation 3 below.

In this case given the average transmit power In order to keep constant P ₂ , g ₀ and P ₁ , It is calculated as in <Equation 4>.

The reallocation of power using Equation 2 will be referred to as generalized power reallocation (GPR). Where P ₁ and P ₂ are determined to be greater than or equal to 0, and include the following special cases:

i) P ₁ = 0: Power Truncation Only (PTO)

ii) P ₂ = 0: Truncated Channel Inversion (TCI)

Since the simple power cut method requires only 1 bit of feedback information, there is an advantage that it can be relatively simple to implement than other reallocation methods.

In another embodiment of the present invention, the power P (g) reassigned to the code symbol is differentiated according to whether the code symbol is a systematic part or a parity part. This reallocation is referred to as two-level power reallocation (GPR-2 level).

In this case, the power P (g) is defined as <Equation 5> and <Equation 6> by applying different power cutoff thresholds g _{0, S} and g _{0, P} for the systematic part and the parity part, respectively. do. In the following, P _S (g) means power allocated to a code symbol corresponding to the systematic part, and P _P (g) means power allocated to a code symbol corresponding to the parity part.

Here, P _{1, S} , P _{2, S} , P _{1, P} , P _{2, P} are all greater than or equal to 0 and include the following special cases.

i) P _{1, S} = P _{1, P} = 0: PTO-2 level

ii) P _{2, S} = P _{2, P} = 0: 2 level power cut channel inversion (TCI-2 level)

g _{0, S} and g _{0, P} , P _{1, S} , P _{2, S} , P _{1, P} , P _{2, P} are similarly shown in Equation 2 above. After modeling, it is set to an optimal value in terms of minimizing the probability of error after decoding sign symbols that have experienced such a channel. In most cases, since the values of g _{0, S} are optimized to be less than or equal to g _{0, P} , the probability that power is not allocated in the parity portion is greater than in the systematic portion. This is because the systematic part of the turbo or LDPC code is more important in minimizing the probability of bit error than the parity part. Here, when the code rate of the error correction code is R in the Rayleigh fading channel, the average transmit power becomes as shown in Equation 7 below.

On the other hand, orthogonal frequency division multiplexing (hereinafter, referred to as OFDM) technology is widely recognized as a core transmission scheme for next generation wireless communication. The reason is that it is simple to implement and has a parallel structure that transmits a high-speed data signal through each frequency known as sub-carrier frequencies (sub-carriers) to support high data rates, and multi-channel This is because it has robustness against fading.

The OFDM scheme transmits low-speed data in parallel on a plurality of carriers instead of high-speed transmission of a single carrier, thereby making it less susceptible to frequency selective fading or narrowband interference. The OFDM scheme has good spectral efficiency because the spectra of subchannels overlap each other while maintaining mutual orthogonality.

In an OFDM system, a transmitted signal is modulated by an Inverse Fast Fourier Transform (IFFT), and a received signal is demodulated by a Fast Fourier Transform (FFT). In this way, an efficient configuration of the digital modulation and demodulation section is possible. The biggest advantage of this configuration is that the receiver can be simply configured as a one-tap equalizer that requires only one complex multiplication for each carrier.

In this OFDM scheme, power is allocated separately by considering N subchannels formed by N subcarriers independently.

3 illustrates a transmission and reception structure of an OFDM mobile communication system according to another embodiment of the present invention.

Referring to FIG. 3, the information sources 10 generate and output an information sequence to be transmitted to the receiver of the opposite party. The channel encoder 12 encodes the input information sequence by a turbo code or an LDPC code according to a predetermined code rate and outputs binary code symbols.

The code symbols are interleaved by a channel interleaver 14 according to a predetermined interleaving rule, and are divided into N paths corresponding to N subcarriers by a serial to parallel converter (S / P) 110. Is distributed. N power allocators 122, 124, and 126 depend on the channel power gains g ₁ , g ₂ , ... g _N of the subcarrier that the corresponding code symbol will experience for the N distributed code symbols. Assign the corresponding powers P ₁ (g ₁ ), P ₂ (g ₂ ), ... P _N (g _N ).

An inverse fast Fourier transformer (IFFT) 130 converts the N code symbols from the power allocators 122, 124, and 126 into N corresponding subcarriers into a signal in the time domain. The converted signal is transmitted to the receiving end via the channel 200. The symbols received at the receiver are g ' ₁ P ₁ (g ₁ ), g' ₂ P ₂ (g ₂ ), ... g ' _N P _N (g _N ) according to the actual channel gain g' of the channel 200. It has a received power expressed by.

The fast Fourier transformer (FFT) 310 converts the received time-domain signal to an N-point fast Fourier transform to output N code symbols corresponding to the N subcarriers, and a parallel to serial converter (P / S) 320 connects the N code symbols in a single path.

The channel deinterleaver 32 returns the code symbols output from the parallel / serial converter 320 in the original order according to the deinterleaving rule corresponding to the interleaving rule used in the channel interleaver 14. The channel decoder 34 decodes the deinterleaved code symbols from the channel deinterleaver 32 according to an error correcting code used in the channel encoder 12, that is, a turbo code or an LDPC code. )

In this case, the power allocators 122, 124, and 126 determine the power P _n (g _n ) allocated to the nth subcarrier according to the channel power gain g _n of the nth subcarrier according to Equation 2 mentioned above. do. In order for the channel power gains g ₁ , g ₂ , ... g _N of the _N subchannels to be independent and equally distributed, the interleaving magnitudes of the channel interleaver 14 and the channel deinterleaver 32 are determined by the coherence of the channel. It is designed to be much larger than the total number of sign symbols transmitted on the N subcarriers during the run time.

Hereinafter, the performance of a turbo code using independent and equally distributed channel power gains according to the present invention will be described.

The turbo code used was a block length of 320, a code rate of 1/3, and the same coder as that of the 3GPP (3rd Generation Partnership Project) design specification, which was decoded by the Max-Log-MAP algorithm. In addition, the channel is a channel with Additive White Gaussian Noise (AWGN) in Rayleigh fading, and coded binary code symbols are modulated with BPSK.

FIG. 4 shows a bit error rate (BER) for noise versus bit energy (Eb / No) when eight iterative decoding is performed on a turbo code. Here, "equal power allocation" is the case of allocating the same power to the code symbols, PTO is the case of simple power-off, PTO-2 level is the case of two-level simple power-off, TCI is the TCI-2 level is the case of 2 level power-off channel reversal, proposed is the case of GPR where both P ₁ and P ₂ are not 0, and proposed-2 level is P _{1, S} , P _{1, P} and P _{2, S} , P _{2, P} are the simulation results in the case of two-level GPR where all are not zero.

As shown, the two-level GPR method according to the present invention can obtain a maximum transmission power gain of 0.86 dB when the BER is 10 ^-6 compared to the method of allocating the same power to all the code symbols. In this case, the result indicated by (A) indicates that the code symbol is transmitted only when the channel condition is good without transmitting the code symbol when the channel condition is bad. The code rate is 1/2 and the number of subchannels is 512. It shows the result of applying Chow algorithm that assigns code symbols adaptively according to channel transmission characteristics (SNR, etc.).

Here, the Chow algorithm is an optimization problem of minimizing the total transmission power under the condition that the total transmission rate is constant and the desired bit error probability is satisfied based on the number of bits and power transmitted for each subchannel based on the SNR of each channel. In solving, solving such an optimization problem requires very high computational complexity. In addition, the SNR of each channel changes after each channel coherence time, so the system needs to be solved again.

Next, look at the performance of the LDPC code of the present invention through the simulation.

The LDPC code used is a regular LDPC code with infinite block length, 1/4 code rate, 3 in the parity check matrix, 3 in the column, and 6 in the row. In consideration of the (erasure), it was decoded with a hard decision based decoding algorithm consisting of three states (-1, 0, 1). In addition, the channel is a channel having AWGN in Rayleigh fading, and coded binary code symbols are modulated with BPSK.

FIG. 5 shows a bit error probability (BER) for noise to bit energy (Eb / No) when infinite repetitive decoding is performed on an LDPC code. Where (a) is the case of GPR, (b) is the case of TCI, (c) is the case of PTO, (d) is the case of allocating the same power to code symbols, and PTO is simple power-off In this case, the simulation results are shown.

As shown, the two-level GPR scheme according to the present invention can obtain a transmission power gain of 1.74 dB compared to the scheme of allocating the same power to all code symbols.

Meanwhile, in the detailed description of the present invention, specific embodiments have been described, but various modifications are possible without departing from the scope of the present invention. Therefore, the scope of the present invention should not be limited to the described embodiments, but should be defined not only by the scope of the following claims, but also by those equivalent to the scope of the claims.

In the present invention operating as described in detail above, the effects obtained by the representative ones of the disclosed inventions will be briefly described as follows.

According to the present invention, high power gain can be obtained through a combined design of power control and error correction code that reallocates and transmits power between code symbols according to channel conditions and decodes them. The present invention can achieve very high power gain for turbo and LDPC codes that provide high coding gain, and provide high power gain even in a simple PTO scheme. As a result, the present invention can obtain the effects of reducing transmission power, reducing interference amount, and extending battery life.

1 is a diagram illustrating a transmission and reception structure of a next generation mobile communication system according to an embodiment of the present invention.

2 is a flow diagram illustrating the operation of allocating power for a sign symbol in accordance with one embodiment of the present invention.

3 is a diagram illustrating a transmission and reception structure of an OFDM mobile communication system according to another embodiment of the present invention.

4 is a diagram showing bit error probability (BER) for noise to bit energy (Eb / No) when eight times of decoding is performed on a turbo code.

5 is a diagram showing bit error probability (BER) for noise to bit energy (Eb / No) when infinite repetitive decoding is performed on an LDPC code.

Claims (10)

A method of power control of a code symbol encoded by an error correction code in a mobile communication system, Obtaining information about an estimated channel power gain for a channel to which a code symbol coded by an error correction code is to be transmitted; Allocating power determined according to the channel power gain to the code symbol when the channel power gain is equal to or greater than a predetermined power cutoff threshold value; And allocating zero power to the code symbol if the channel power gain is not greater than or equal to the power off threshold. The method as claimed in claim 1, wherein the error correction code is a turbo code or a Low Density Parity Check (LDPC) code. 3. The method of claim 2, wherein if the channel power gain is greater than or equal to a predetermined power off threshold, the power is greater than or equal to zero after dividing a predetermined first value by the channel power gain to be greater than or equal to zero. And a predetermined second value is determined as a value. A method of power control of a code symbol encoded by an error correction code in a mobile communication system, Obtaining information about an estimated channel power gain for a channel to which a code symbol coded by an error correction code is to be transmitted; Determining whether the code symbol is a systematic part or a parity part; The channel power gain for the code symbol if the channel power gain is equal to or greater than a first power cutoff threshold for a predetermined systematic portion or a second power cutoff threshold for a predetermined parity portion according to the determination result. Allocating the power determined according to, And allocating zero power to the code symbol if the channel power gain is not greater than the first or second power off threshold. The method as claimed in claim 4, wherein the error correction code is a turbo code or a Low Density Parity Check (LDPC) code. 6. The method of claim 5, wherein if the code symbol corresponds to a systematic part and the channel power gain is greater than or equal to the first power cutoff threshold, the power is set to P _{1, S} + P _{2, S} / g, Where P _{1, S} and P _{2, S} are predetermined values for power control of the systematic part and g is the channel power gain, If the code symbol corresponds to a parity portion and the channel power gain is greater than or equal to the second power cutoff threshold, then the power is defined as P _{1, P} + P _{2, P} / g, where P _{1, P} and P _{2 , P} are predetermined values for power control of the systematic part. An apparatus for power control of a code symbol encoded by an error correction code in a mobile communication system, A channel encoder for generating code symbols by encoding an input information sequence according to a predetermined code rate; If the estimated channel power gain for the channel to which the code symbol is to be transmitted is equal to or greater than a predetermined power cutoff threshold value, power determined according to the channel power gain is allocated to the code symbol, and the channel power gain is the power cutoff threshold value. And a power allocator for allocating zero power to the code symbol. 8. The apparatus as claimed in claim 7, wherein the error correction code is a turbo code or a Low Density Parity Check (LDPC) code. The method of claim 8, wherein the power allocator, If the channel power gain is equal to or greater than a predetermined power cutoff threshold value, the power to be allocated to the code symbol is determined by dividing a first predetermined value by the channel power gain and adding a second predetermined value. The device, characterized in that. The method of claim 8, wherein the power allocator, Determining whether the code symbol is a systematic part or a parity part, If the code symbol corresponds to the systematic portion and the channel power gain is greater than or equal to the first power cutoff threshold, the power is set to P _{1, S} + P _{2, S} / g, where P _{1, S} and P _{2, S} are predetermined values for power control of the systematic part and g is the channel power gain, If the code symbol corresponds to a parity portion and the channel power gain is greater than or equal to the second power cutoff threshold, then the power is defined as P _{1, P} + P _{2, P} / g, where P _{1, P} and P _{2 , P} are predetermined values for power control of the systematic part.