RU2739940C1 - Adaptive modulation method for communication systems using orthogonal frequency multiplexing signals - Google Patents

Abstract

FIELD: radio equipment.

SUBSTANCE: invention relates to radio engineering, particularly to adaptive modulation techniques, which can be used in wireless and wire communication systems using orthogonal frequency multiplexing (OFDM) signals for transmission. Transmitting side generates a pilot OFDM symbol in each transmitted frame. At the receiving side, after synchronization and equalization, the error vector is calculated between the received points of the constellation and the initial value of each point of the pilot OFDM symbol. Further, maximum value of error vector is selected for each of subcarrier subgroups. Next step is to correct the obtained error vector values for subcarrier subgroups based on previous measurements, which consists in storing values of previous estimation of error vectors, comparing them with the new estimate and making a decision on the steady-state evaluation value based on the comparison and assigning the modulation indices in accordance with the set thresholds selected to provide the given probability of the bit error in the receiving device.

EFFECT: technical result consists in reducing the number of bit errors when transmitting signals in channels with active interference, with low SNR, as well as in channels with multi-beam, non-constant, rapidly changing characteristics.

1 cl, 3 dwg

Description

The invention relates to radio engineering, in particular to methods of adaptive modulation, which can be used in wireless and wired communication systems using orthogonal frequency division multiplexing (OFDM) signals for transmission.

The known method of adaptive modulation, described in the article [1]. This method consists in the fact that the ratio of the average power of the received useful signal to the noise power (SNR) is calculated from the received signal after time synchronization. Next, according to the obtained SNR value, the most suitable modulation index is selected. Moreover, SNR is calculated from a random signal, pilot sequences are not used. The SNR in this method is estimated by measuring the vector modulus between the received constellation points and the nearest reference points, the calculated values are compared with the threshold values, after which the type of modulation is selected. The article gives thresholds for the following modulation indices: BPSK, QPSK, 16-QAM, 64-QAM, 256-QAM. Further, the resulting estimate is transmitted to the transmitter, where it is used in modulating subsequent data streams.

The disadvantage of this method is the increased probability of a system bit error at low SNR. Since only a random received signal that is not known in advance by the receiver is used to estimate the SNR, the pilot sequence is not used. In this regard, in the case of a low SNR in the receiver, the assessment will be incorrect, since there is no information on the initial position of the received constellation point on the receiving side. An incorrect SNR estimate occurs due to the fact that a point can move from the area of the original control point to the area of the adjacent constellation control point. This, in turn, will lead to an error in the SNR estimate, since the SNR estimate in this method is made by measuring the vector modulus between the received constellation points and the nearest control points, and not the original ones.

Closest to the claimed method of adaptive modulation is the method given in the article [2]. In this method, a signal is received, after which time synchronization, evaluation of the communication channel and equalization are carried out. Next, the signal spectrum is divided into subgroups of subcarriers, for which the signal-to-noise ratio is calculated for each subcarrier and the average signal-to-noise ratio for the subgroup. After the listed calculations, a modulation index is assigned in each subgroup in accordance with the threshold modulation grid (thresholds for the types of modulations used). The next step is to calculate the total number of bits per OFDM symbol. Next, an array of assigned modulation indices is transmitted to the transmitting part via the most noise-immune communication channel. Then the subsequent data is modulated in accordance with the received array.

The disadvantage of this method is the increased probability of a system bit error at low SNR. Since only a random received signal that is not known in advance by the receiver is used to estimate the SNR, the pilot sequence is not used. In this regard, in the case of a low SNR in the receiver, the SNR estimate will be performed incorrectly, since there is no information on the initial position of the received constellation point on the receiving side. An incorrect SNR estimate occurs due to the fact that a point can move from the area of the original control point to the area of the adjacent constellation control point. This, in turn, will lead to an error in the SNR estimate, since the SNR estimate in this method is made by measuring the vector modulus between the received constellation points and the nearest control points, and not the original ones. In addition, the disadvantage of this method is that the decision on the assignment of the modulation index for a group of subcarriers is made on the basis of a single estimate, since the frequency interval of the location of the group of subcarriers may overlap with periodically occurring active interference, which may not act at the time of assessment, but occur during subsequent transmission data. This will also lead to additional errors.

The problem to be solved by the proposed technical solution is to reduce the number of errors in signal transmission in channels with active interference, at low SNR, as well as in channels with multipath, non-constant, rapidly changing characteristics.

The solution to this problem is achieved by the fact that in the adaptive modulation method, including the reception of the signal generated by the transmitter, synchronization in time and frequency, calculation of the fast Fourier transform for information OFDM symbols, estimation of the channel transfer characteristic, equalization, dividing the signal spectrum into subgroups of subcarriers, transmission of an array of indices modulation on the additional noise-immune channel, the operation of generating a pilot OFDM symbol in each transmitted frame is additionally performed, receiving a pilot OFDM symbol, calculating the fast Fourier transform for the pilot OFDM symbol, estimating the channel transfer characteristic from the pilot OFDM symbol, equalizing the pilot OFDM symbol, calculating the error vector between the received constellation points and the initial value of each point of the pilot OFDM symbol, selection of the maximum value of the error vector for each of the subcarrier subgroups, filtering (correction) of the received values of the error vectors for subgroups of subcarriers based on previous measurements, including storing the values of the previous estimate of the error vectors, comparing them with the new estimate, deciding on the steady-state value of the estimate based on the comparison, assigning modulation indices in accordance with the set thresholds selected to provide a given probability of a bit error in the receiving device ...

The functional diagram of the proposed method is shown in Fig. 1, which indicates: 1 - Reception of the n-th pilot OFDM symbol, 2 - Time synchronization 3 - Frequency synchronization, 4 - Fast Fourier transform, 5 - Propagation channel estimate for pilot subcarriers, 6 - Equalization, 7 - Error vector estimate for each subcarrier (e _n ), 8 - Grouping m subgroups by l subcarriers of the signal spectrum (E _n ), 9 - Selecting the maximum value of the error vector in each i-th subcarrier subgroup (M _n ), 10 - Filtering the obtained values based on previous measurements (R _n ), 11 - Assignment of modulation indices by subgroups (P _n ), 12 - Transfer of the received array of modulation indices to the transmitting part.

Detailed description of the method

On the transmitting side, a pilot OFDM symbol is formed, having a structure similar to the information symbols following it. To generate the pilot OFDM symbol, a sequence of bits is used, which is modulated by a quadrature amplitude modulator (QAM). The thus obtained pilot OFDM symbol is added at the beginning of each frame according to the structure shown in FIG. 2. The frame consists of a pilot OFDM symbol P and t information symbols U _t . The pilot symbol does not carry user information and is fully known on the receiving side.

The receiver receives and digitizes the frame. Time synchronization is performed with the beginning of the frame on the pilot OFDM symbol. Time synchronization can be performed by calculating the cross-correlation function between the received pilot OFDM symbol and the reference signal [3]. Next, frequency synchronization is performed. Frequency synchronization algorithms are given in [4]. To move from the time domain to the frequency domain, a fast Fourier transform is calculated. Then, the transfer function of the communication channel is estimated from the pilot subcarriers. The obtained estimate is used for the operation of equalizing the spectrum of the received pilot OFDM symbol [5].

:Spectrum samples of the received OFDM symbol represent a vector

:

, (1)

, (one)

Where:

- это номер принятого пилотного OFDM символа,

is the number of the received pilot OFDM symbol,

- это номер поднесущей в спектре пилотного OFDM символа,

is the subcarrier number in the spectrum of the pilot OFDM symbol,

- это общее количество используемых поднесущих спектра пилотного OFDM символа.

is the total number of used spectrum subcarriers of the pilot OFDM symbol.

в частотной области:The reference signal in the receiving part is a vector of reference values

in the frequency domain:

. (2)

... (2)

для каждой поднесущей по следующей формуле (3):Then the error vectors are calculated

for each subcarrier according to the following formula (3):

. (3)

... (3)

в

подгрупп по

поднесущих. Данная группировка позволяет сократить количество необходимых вычислений, выполняемых в дальнейшем, и уменьшить количество служебной информации, требуемое для передачи массива индексов модуляций на передающее устройство. Количество поднесущих в одной подгруппе, как и само количество подгрупп, определяется полосой когерентности канала.The next step is to group the error vectors

in

subgroups on

subcarriers. This grouping allows you to reduce the number of necessary calculations performed in the future, and to reduce the amount of service information required to transfer the array of modulation indices to the transmitting device. The number of subcarriers in one subgroup, as well as the number of subgroups itself, is determined by the channel coherence band.

. (4)

... (4)

:Further, in each subgroup, the maximum value is selected, thus obtaining the vector of the maximum error in each of the subgroups of subcarriers

:

. (5)

... (5)

The maximum error vector value corresponds to the worst case in the sub-carrier subset. This estimation method is supposed to be used in channels with multipath signal propagation and variable, rapidly changing characteristics. In this case, the signal propagation channel will have a small coherence band, which will include a small number of subcarriers. As a practical study of the algorithm has shown, the greatest reliability and stability in this case is provided by the choice of the maximum value in the subgroup, and not the average.

. Предполагается, что в приемном устройстве хранится предыдущие значения оценки. Суть данной операции описывается формулой (6), представленной ниже.The next stage is filtering and adjusting the resulting array of error vectors

... It is assumed that the receiver stores the previous evaluation values. The essence of this operation is described by formula (6) presented below.

. (6)

... (6)

Where:

- это итоговая оценка канала передачи на текущий момент времени,

is the final estimate of the transmission channel at the current time,

- это итоговая оценка канала передачи, выполненная в предыдущую итерацию алгоритма,

is the final estimate of the transmission channel made in the previous iteration of the algorithm,

- это декремент фильтра.

is the decrement of the filter.

больше установленной ранее

принимается текущее значение оценки вектора ошибки. Если ситуация в канале улучшилась, то есть значение нынешней оценки вектора ошибки

меньше установленной ранее

то от предыдущей оценки отнимается значение

которое вычисляется по формуле (7):If the situation has worsened for a given subgroup of subcarriers, that is, the value of the current estimate of the error vector

more established earlier

the current value of the error vector estimate is taken. If the situation in the channel has improved, that is, the value of the current estimate of the error vector

less than previously established

then the value

which is calculated by the formula (7):

. (7)

... (7)

Where:

- коэффициент инерционности фильтра.

- coefficient of inertia of the filter.

определяет инерционность фильтра. Чем больше данный коэффициент, тем медленнее реакция фильтра на улучшение в канале. Значение коэффициента выбирается исходя из времени когерентности канала.Thus, we get a filter that responds quickly to impairments in the transmission channel and slowly to improvements. This allows for a more stable estimate of the channel interference environment as it is less affected by the random improvements that can occur in sub-groups of subcarriers as a result of a single measurement giving an estimate indicative of improvements in the transmission channel. This approach allows you to keep the bit error probability at the same level and does not lead to a sharp increase in the number of errors. Coefficient

determines the inertia of the filter. The larger this coefficient, the slower the filter's response to an improvement in the channel. The coefficient value is chosen based on the channel coherence time.

является итоговой оценкой канала передачи на текущий момент времени и используется для назначения порогов. Пороги выбираются исходя из требуемой вероятности битовой ошибки на приемном устройстве. На фиг. 3 приведена зависимости вероятности битовой ошибки от значения порога для некоторых видов модуляции, на котором обозначено: зависимость вероятности битовой ошибки от порога

для BPSK, QPSK, QAM16, QAM64. Массив номеров индексов модуляций определяется исходя из условий, представленных в формуле (8):The resulting array

is the final estimate of the transmission channel at the current time and is used to assign thresholds. The thresholds are selected based on the required bit error probability at the receiver. FIG. 3 shows the dependence of the bit error probability on the threshold value for some types of modulation, on which it is indicated: the dependence of the bit error probability on the threshold

for BPSK, QPSK, QAM16, QAM64. The array of numbers of modulation indices is determined based on the conditions presented in the formula (8):

. (8)

... (8)

In the presented example, the numbering of modulation types goes from 1 to 4 in ascending order of the modulation index. The value "0" corresponds to the ranges of subcarriers that are prohibited for data transmission.

After the assignment of the modulation indices is performed, the resulting array must be sent to the transmitting side via a highly reliable, more noise-immune service data channel. In addition, this array enters the information data receiving unit for correct demodulation of subsequent messages.

The proposed method provides a reduction in the number of errors in signal transmission at low SNR, as well as in channels with multipath and variable, rapidly changing characteristics. The advantage of this method is that using the pilot sequence to perform the transmission channel estimate improves the accuracy of its determination. The procedure of additional filtering of estimates allows to take into account previous estimates, quickly respond to impairments in the transmission channel (the occurrence of impulse and narrowband interference) and slowly to improve. This eliminates the possibility of making an incorrect estimate in the case of random (or very short-term) improvements in the channel, which usually lead to a large number of errors on the receiving side. This makes it possible to provide more stable communication with the given characteristics in terms of the bit error probability on the receiving side. Reducing the probability of a bit error can reach 10% compared to the prototype method.

Used sources

1. Rajesh M.N. et al. An analysis of BER comparison of various digital modulation schemes used for adaptive modulation // 2016 IEEE International Conference on Recent Trends in Electronics, Information & Communication Technology (RTEICT). - IEEE, 2016 .-- S. 241-245.

2. Grunheid R., Bolinth E., Rohling H. A blockwise loading algorithm for the adaptive modulation technique in OFDM systems // IEEE 54 ^th Vehicular Technology Conference. VTC Fall 2001. Proceedings (Cat. No. 01CH37211). - IEEE, 2001 .-- T. 2. - S. 948-951.

3. Pat. RF No. 2367972, IPC G01S 5/06. A method for assessing the accuracy of determining the location of a radio emission source by a passive differential-rangefinder system. Publ. 20.09.2009.

4. Shakhtarin B.I. et al. Method of frequency synchronization for OFDM systems in channels with additive white Gaussian noise and Rayleigh fading. Bulletin of the Moscow State Technical University. NE Bauman. Instrument making series. - 2016. - No. 2 (107).

5. Rogozhnikov E.V., Abenov R.R., Vershinin A.S., Voroshilin E.P. Investigation of equalization methods for communication systems using OFDM signals // Bulletin of SibGUTI. 2013. No. 1 (21). S. 50-56.

Claims (1)

An adaptive modulation method for communication systems using signals with orthogonal frequency multiplexing, including receiving the signal generated by the transmitter in the form of information OFDM symbols transmitted over frames, synchronization in time and frequency, calculating the fast Fourier transform for information OFDM symbols with the formation of a constellation, estimating the transfer characteristic channel, equalization, dividing the signal spectrum into subgroups of subcarriers, transmitting an array of modulation indices over an additional noise-immune channel, characterized in that in each transmitted frame an operation of generating a pilot OFDM symbol is additionally performed, having a structure similar to the following information OFDM symbols, receiving a pilot OFDM symbol, calculating the fast Fourier transform for the pilot OFDM symbol, estimating the channel transfer characteristic from the pilot OFDM symbol, equalizing the pilot OFDM symbol, calculating the error vector between the received constellation of information symbols and the initial value of each point of the pilot OFDM symbol, selection of the maximum value of the error vector for each of the subcarrier subgroups, filtering and correction of the received error vector values for subcarrier subgroups based on previous measurements, including storing the values of the previous estimate of the error vectors, comparing them with by a new estimate, deciding on a steady-state value of the estimate based on the comparison used to assign thresholds, assigning modulation indices in accordance with the set thresholds selected to provide a given bit error probability in the receiver.

Images