CN101005475A

CN101005475A - Method and system for synchronizing time and frequency in orthogonal frequency division multiplex communication

Info

Publication number: CN101005475A
Application number: CN 200610157618
Authority: CN
Inventors: 刘田; 唐友喜; 邵士海; 王吉滨; 李云岗
Original assignee: Huawei Technologies Co Ltd; University of Electronic Science and Technology of China
Current assignee: Huawei Technologies Co Ltd; University of Electronic Science and Technology of China
Priority date: 2006-12-14
Filing date: 2006-12-14
Publication date: 2007-07-25

Abstract

In the invention, at transmitting terminal, the original information treated with carrier modulation is serial-parallel converted into several branch data streams whose amounts are identical with the amounts of antennas; the branch data streams and frequency-domain pilot frequency sequence are commonly mapped into the data position corresponding to the preset frequency-domain pilot frequency pattern; generating each antenna data information which is processed with OFDM modulation and is transmitted by RF transmission. At receiving terminal, each antenna receives the RF signals and generates time-domain output sequence the time-domain output sequence is correlated with the reference sequence to construct synchronous objective function between each pair of antennas so as to implement the synchronization between time and frequency.

Description

Method and system for time and frequency synchronization in orthogonal frequency division multiplexing communication

Technical Field

The present invention relates to multiplexed communications, and more particularly, to a method and system for time and frequency synchronization in Orthogonal Frequency Division Multiplexing (OFDM) communications.

Background

MIMO (multiple-Input multiple-Output) and OFDM (orthogonal frequency Division Multiplexing) technologies are two important items of modern communication technologies. The OFDM technology, as an efficient multi-carrier transmission technology, has the advantages of high spectrum utilization rate and strong frequency selective fading resistance. The MIMO technology can provide a larger channel capacity than the SISO (Single-Input Single-Output) technology, and is a potential method for high-speed data transmission. The MIMO-OFDM method combining MIMO and OFDM can achieve the purpose of providing a high data transmission rate service in a broadband wireless communication system. Since the OFDM system has high requirements for time and frequency synchronization, especially strict requirements for frequency synchronization, and a small frequency offset will cause a large performance loss, time and frequency synchronization is very important in MIMO-OFDM.

The OFDM time and frequency synchronization method is classified into a data-aided synchronization method and a blind synchronization (i.e., synchronization without data assistance) method, and the present invention relates to a data-aided synchronization method. The data-aided synchronization method requires the insertion of a time domain training sequence or a frequency domain pilot in a signal, and thus reduces data transmission efficiency compared to the blind synchronization method, but the data-aided synchronization method has high estimation accuracy and low computational complexity.

Currently available OFDM data-assisted synchronization methods include time and frequency synchronization using one or more OFDM symbols as training sequences, and frequency synchronization using OFDM symbols with inserted pilots. The frequency synchronization is carried out by using the OFDM symbols with the inserted pilot frequency, namely, the same pilot frequency sequence is inserted into the same subcarrier position of the adjacent OFDM symbols, and the carrier frequency offset is estimated by using the differential correlation of the same pilot frequency sequence under the condition that the accurate time synchronization is finished.

Setting the position and content of the training sequence can be used for obtaining better synchronization performance, for example, a sandwich mode is used for placing the training sequence at two ends of a plurality of data symbols, and the training sequence has higher precision when being used for frequency synchronization because the training sequence has longer interval; a repetitive arrangement of multiple CAZAC (Constant Amplitude Auto-Correlation) sequences can be used as the content of the training sequence for frequency synchronization.

Currently, the research on the synchronization of the MIMO-OFDM time and frequency is still in the initial stage, and mainly performed in the following two ways: one is the way in which each antenna transmits multiple training sequences, which is based on the assumption that the time delay of the signal from all transmit antennas to all receive antennas is the same, and the frequency offset is also the same. Performing DFT (Discrete Fourier Transform) on a received signal sequence, then performing correlation with a DFT form of a local time domain sequence to complete frequency coarse synchronization, performing delay correlation modulo on the received signal to realize time coarse synchronization, performing delay correlation modulo to realize frequency fine synchronization, and performing correlation completion on the received signal and the local sequence; the other is a mode that each antenna sends a training sequence staggered in the time domain, the training sequence of the mode still consists of a plurality of OFDM symbols, the time coarse synchronization is realized by taking a delay correlation module and accumulating by a received signal, the time fine synchronization is realized by taking a correlation module by the received signal and a local sequence, and the frequency synchronization is completed by utilizing phase information obtained by taking the delay correlation by the received signal.

It should be noted here that, in the OFDM time and frequency synchronization method, the local sequence and the training sequence correspond to the same mathematical entity, and the local sequence and the training sequence also have the same expression form in the time domain and the frequency domain, respectively. In the technical literature in general, the mathematical entity is called training sequence at the transmitting end; at the receiving end, it is called a local sequence to distinguish it from the received signal sequence. The representation is also employed in the present disclosure.

Two time and frequency synchronization methods of the prior art are as follows:

the first is a method of superimposing PN (Pseudo Noise) sequences. The structure of the training sequence for synchronization is shown in fig. 1, where N is the number of OFDM subcarriers, the period of the PN sequence is K, and L is an integer part of N/K. As can be seen from the figure, the last PN sequence may not be complete. The purpose of the repeated placement of the multiple PN sequences is to correlate the received signal with the local sequence at the receiving end by using the periodic correlation property of the PN sequences, so that a PN sequence having a good autocorrelation property, such as an M sequence or a Gold sequence, can be selected.

At the receiving end, the receiving sequence is correlated with the local sequence to obtain the following correlation function:

in formula (1), c (k) is a training sequence composed of a repeated PN sequence, and r (k) is a sequence obtained by a/D (analog/digital) conversion of a received signal. And P is a positive integer and is used for adjusting the frequency offset estimation range and precision. The objective function of time synchronization is taken as gamma (k, a). And (3) adopting a Sliding Window (Sliding Window) with the length of KL at a receiving end, processing data in the Sliding Window according to the formula (1), wherein a represents the distance of the Sliding Window relative to the precise synchronization point, and a is 0 to represent that time synchronization is finished. For example: assuming that each parameter value is N-128, K-15, and P-1, a total of 8 complete PN sequences are included in the training sequence. When the time delay is 128 sampling intervals, with the movement of the sliding window, there is a small peak output when the local sequence c (k) is aligned with 2 PN sequences, a maximum peak output when the local sequence c (k) is aligned with 8 PN sequences, and the output value is small when the PN sequence is not aligned with the local sequence, so that it is easy to find the time synchronization point. The time synchronization method can complete coarse time synchronization and fine time synchronization at one time, and has high capture probability and low false alarm probability and false alarm probability.

In actual transmission, a training sequence is superimposed on the data at a power ratio, as shown in fig. 2, where the ratio of training sequence to data power is ρ: (1- ρ). According to the working principle of direct sequence spread spectrum, when the PN sequence is correlated, the interference of part of data to the PN sequence can be inhibited.

The frequency synchronization method is to normalize the target function | gamma (k, a) | of time synchronization and take the phase after obtaining the time synchronization to obtain

Wherein sigma_s ²For the power of the received useful signal, γ (k, 0) indicates that time synchronization has been achieved accurately, i.e., a is 0.

The main disadvantages of the method for superimposing PN sequences are: the power of the PN sequence defines the minimum value of the error rate, generally, the transmission signal requires the error rate as low as possible, and the time and frequency synchronization requires the PN sequence to have certain power, which are contradictory; the PN sequence is superposed on random data and can only be used for synchronization, and additional overhead is also needed for channel estimation and the like to realize the correct demodulation of the signal; the method is originally aimed at a SISO system, if the method is directly applied to a MIMO-OFDM system, each antenna needs to be superposed with a training sequence, under the objective condition that the total transmitting power is certain, the proportion of the data power to the total power is further reduced, and the detection of the data is further interfered by a PN sequence.

The second method in the prior art is a method for time synchronization of a pilot-assisted OFDM system, which uses known frequency-domain pilot symbols for channel estimation in combination with a cyclic prefix to complete time synchronization. The specific process is as follows:

let an OFDM symbol comprise N subcarriers, where N is_pKnown pilot symbols are inserted into the sub-carriers, and gamma represents the position sequence number set of the pilot sub-carriers, so that the transmission signal can be regarded as the sum of two parts of data and pilot. The first part being N-N_pThe corresponding time domain signal of each data subcarrier is:

<math> <mrow> <mi>s</mi> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mn>1</mn> <msqrt> <mi>N</mi> </msqrt> </mfrac> <munder> <mi>Σ</mi> <mrow> <mi>k</mi> <mo>&Element;</mo> <mo>{</mo> <mn>0</mn> <mo>,</mo> <mi>KN</mi> <mo>-</mo> <mn>1</mn> <mo>}</mo> <mo>\</mo> <mi>γ</mi> </mrow> </munder> <msub> <mi>x</mi> <mi>k</mi> </msub> <msup> <mi>e</mi> <mrow> <mi>j</mi> <mn>2</mn> <mi>πkn</mi> <mo>/</mo> <mi>N</mi> </mrow> </msup> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>3</mn> <mo>)</mo> </mrow> </mrow> </math>

where {0, K N-1} \ γ denotes the difference set of the set {0, K N-1} and the set γ, x_kIs the data symbol transmitted on the k sub-carrier, x_kCan be mapped in any known constellation mapping manner and has an average desired power

. The second part comprises N_pA pilot subcarrier corresponding to the time domain signal of

<math> <mrow> <mi>m</mi> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mn>1</mn> <msqrt> <mi>N</mi> </msqrt> </mfrac> <munder> <mi>Σ</mi> <mrow> <mi>k</mi> <mo>&Element;</mo> <mi>γ</mi> </mrow> </munder> <msub> <mi>p</mi> <mi>k</mi> </msub> <msup> <mi>e</mi> <mrow> <mi>j</mi> <mn>2</mn> <mi>πkn</mi> <mo>/</mo> <mi>N</mi> </mrow> </msup> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>4</mn> <mo>)</mo> </mrow> </mrow> </math>

Wherein p is_kIs the pilot symbol transmitted on the k sub-carrier, its expected power

The received signal in an AWGN (Additive White Gaussian Noise) channel can be expressed as

r(n)＝s(n-θ)+m(n-θ)+w(n) (5)

Where θ represents an unknown integer multiple of the normalized time delay, w (n) is additive complex white Gaussian noise, and w (n) has a variance σ_w ². The statistical properties of the received signal r (n) are as follows:

for OFDM systems with a large number of data subcarriers (N)_pN), the time-domain signal s (N) may be approximated as a discrete-time gaussian process with a variance α σ_x ²，α＝(N-N_p) N; and the pilot signal m (n) is a determination signal. Therefore, the received signal r (n) is a Gaussian process with a time-varying mean m (n) and a variance α σ_x ²+σ_w ². Adopting ML algorithm, when the maximum likelihood condition is satisfied, the time delay estimation

The calculation formula of (2) is as follows:

wherein,

Λ(θ)＝ρΛ_cp(θ)+(1-ρ)Λ_p(θ) (7)

the ML algorithms described in equations (5) to (10) are suitable for the case of no frequency offset, and when there is frequency offset, the received signal can be expressed as

r(n)＝(s(n-θ)+m(n-θ))e^j2πεn/N+w(n) (11)

Where ε is the normalized frequency offset, and ε is expressed as the sum of two parts: e ═ e-_I+ε_fWherein, epsilon_IRepresenting an integer part of the frequency offset, ε_fRepresenting fractional part frequency offset.

Certain randomness exists in formula (7) due to changes of frequency offset or channel phase, which results in estimation results

The variance increases and the ML algorithm described above needs to know the receiver SNR (signal to noise ratio) when calculating p, which is often not feasible. Therefore, for the two points, a robust time synchronization algorithm is proposed again:

wherein,

wherein S²NR is an assumed SNR value and is fixed.

It can be seen that the method for time synchronization of the pilot-assisted OFDM system has the following disadvantages: the adopted ML algorithm is derived based on an AWGN channel, and when the ML algorithm is used for a multipath fading channel, the autocorrelation characteristic of a receiving sequence r (k) can show complex change, so that the random change of a likelihood function is caused, and the performance of the algorithm is obviously reduced; the signal-to-noise ratio of the received signal needs to be estimated, and the estimation error can cause the reduction of the synchronization precision; although time synchronization and channel estimation can be performed simultaneously, frequency synchronization needs to be performed additionally, which does not substantially reduce the overhead of the system; more difficult to use in more complex MIMO channel environments.

Disclosure of Invention

The technical problem to be solved by the embodiments of the present invention is to provide a method and a system for synchronizing time and frequency in ofdm communication, which have high efficiency of carrying effective data and small consumption of synchronization process.

In order to solve the technical problems, the technical scheme is as follows:

the embodiment of the invention provides a method for synchronizing time and frequency in orthogonal frequency division multiplexing communication, which comprises the following steps:

original information after carrier modulation is serial-parallel converted into n_TStrip tributary data stream, n_TIs the number of transmit antennas;

constructing a frequency domain pilot frequency sequence of a transmitting antenna;

mapping the branch data stream of the transmitting antenna and the frequency domain pilot frequency sequence to the corresponding data position of a preset frequency domain pilot frequency pattern to generate data information;

the data information is transmitted by a transmitting antenna after OFDM modulation;

the receiving antenna receives electromagnetic waves and generates a time domain output sequence;

constructing a synchronous target function between the antennas according to the correlation between the time domain output sequence and the reference sequence;

and estimating a time synchronization point and a frequency offset by using the peak value information of the synchronization objective function.

And completing the time and frequency synchronization according to the time synchronization point and the frequency offset information.

The embodiment of the invention also provides an orthogonal frequency division multiplexing communication system capable of realizing time and frequency synchronization, wherein a transmitting end of the system comprises: the device comprises a multi-input multi-output encoder, at least one OFDM modulator and a radio frequency transmitting part, wherein the multi-input multi-output encoder is used for encoding original input information and generating data information of each antenna according to a frequency domain pilot frequency pattern;

its receiving end includes: the device comprises a radio frequency receiving part for receiving radio frequency signals and carrying out radio frequency processing to generate time domain output sequences of the received signals, a time and frequency synchronizing device for constructing a synchronizing objective function which takes a time point as an independent variable among each pair of antennas according to the correlation between the time domain output sequences and a local sequence and estimating the time synchronizing point and the frequency offset by using peak value information of the synchronizing objective function, at least one OFDM demodulator for carrying out OFDM demodulation on the time domain output sequences of the received signals by combining the estimated time synchronizing point and frequency offset information to obtain restored data information of each antenna, and a multi-input multi-output decoder for carrying out guide auxiliary channel estimation on the restored data information of each antenna by using frequency domain pilot frequencies and then carrying out space-time frequency decoding to obtain restored original input information.

The method and the system for synchronizing the time and the frequency in the orthogonal frequency division multiplexing communication provided by the embodiment of the invention can obtain the following beneficial effects: by designing special pilot sequences of all transmitting antennas, the system can realize time and frequency synchronization with lower cost, and each antenna can complete a synchronization task only by sending an OFDM symbol with pilot inserted in a frequency domain without additional training sequences.

Drawings

Fig. 1 is a schematic block diagram of an OFDM synchronization training sequence.

Fig. 2 is a schematic diagram of a signal superposition mode of an OFDM synchronous training sequence.

Fig. 3 is a flow chart illustrating a method of time and frequency synchronization in a preferred embodiment of the present invention.

Fig. 4 is a schematic block diagram of an OFDM communication system capable of achieving time and frequency synchronization in a preferred embodiment of the present invention.

Fig. 5 is a schematic diagram of an embodiment of a frequency-domain pilot pattern of the antenna joint design in a preferred embodiment of the present invention.

Fig. 6 is a diagram of a time-domain pilot sequence with a cyclic prefix in a preferred embodiment of the present invention.

Fig. 7 is a schematic block diagram of a MIMO encoder in a preferred embodiment of the present invention.

Fig. 8 is a schematic block diagram of an OFDM modulator in a preferred embodiment of the invention.

Fig. 9 is a functional block diagram of a radio frequency transmitter in accordance with a preferred embodiment of the present invention.

Fig. 10 is a block diagram of the synchronization algorithm of the time and frequency synchronization block in a preferred embodiment of the present invention.

Fig. 11 is a schematic block diagram of a radio frequency receiver in a preferred embodiment of the invention.

Fig. 12 is a schematic block diagram of an OFDM demodulator in a preferred embodiment of the present invention.

Detailed Description

The method for synchronizing the time and the frequency between the transmitting signal and the receiving signal in the OFDM communication can be used for an MIMO system. The method is a joint process including transmit data construction at the transmitting end and time and frequency synchronization at the receiving end with the transmit data of a particular structure. The idea is that the communication process uses pilot frequency, the time domain sequence has specific correlation by designing the frequency domain joint pilot frequency pattern of each antenna of the transmitting terminal, and then the time domain correlation is used for time and frequency synchronization.

The following describes a method and a system for MIMO-OFDM time and frequency synchronization according to an embodiment of the present invention in detail with reference to the accompanying drawings.

In a preferred embodiment of the present invention, a flow chart of the MIMO-OFDM time and frequency synchronization method is shown in fig. 3, in which steps 301 to 304 are steps of a transmitting end, and steps 305 to 307 are steps of a receiving end.

First, in step 301, original information modulated by a carrier is converted into n by serial-parallel conversion_TA stripe tributary data stream, where n_TIs the number of transmit antennas; then, step 302 constructs the frequency domain pilot frequency sequence of each transmitting antenna; then, in step 303, mapping the branch data streams of each transmitting antenna generated in step 301 and the frequency domain pilot sequence constructed in step 302 to corresponding data positions of a predetermined frequency domain pilot pattern together to generate each transmitting antennaThe data information of (2); at the end of the transmitting end, step 304, the data information of each antenna is OFDM modulated and transmitted via radio frequency.

At the signal receiving end, firstly, in step 305, receiving the electromagnetic wave by each receiving antenna to generate a time domain output sequence; then, in step 306, according to the correlation between the time domain output sequence and the local sequence, a synchronization objective function with the time point as an argument between each pair of antennas is constructed; thereafter, the time synchronization point and the frequency offset are estimated using the peak information of the synchronization objective function in step 307.

It should be noted that step 301 and step 302 are not in a bearing relationship, and the order of the two steps may be arbitrary, that is, either step may be executed first, or both steps may be executed simultaneously and separately.

The OFDM communication time and frequency synchronization method according to the embodiment of the present invention is implemented by being applied to an OFDM communication system. In a preferred embodiment of the present invention, a schematic block diagram of a system using the MIMO-OFDM communication time and frequency synchronization method is shown in fig. 4. The MIMO-OFDM communication system includes a space-time-frequency modulation section 401, a radio frequency transmission section 406, a radio frequency reception section 413, a time-frequency synchronization device 420, and a space-time-frequency demodulation section 421.

Wherein, the space-time-frequency modulation part 401 and the radio frequency transmitting part 406 are positioned at a transmitting end. The space-time-frequency modulation section 401 further includes: a MIMO encoder 402 for encoding the original input information and generating data information of each antenna according to the frequency domain pilot pattern, wherein a more detailed internal structure of the MIMO encoder 402 will be described in detail below with reference to fig. 6; a plurality of OFDM modulators (403, 404, 405) that perform OFDM modulation on data information of each antenna to generate OFDM symbols, wherein the OFDM modulation can be realized by IDFT (Inverse Discrete Fourier Transform); and the radio frequency transmitting part comprises a plurality of groups of combinations of radio frequency processing I (407, 408, 409) and transmitting antennas (410, 411, 412), each group of combinations (such as the combination of the radio frequency processing I407 and the transmitting antennas 410) corresponds to one OFDM modulator, and the OFDM symbols are subjected to radio frequency processing by the radio frequency processing I and are transmitted through the transmitting antennas.

The radio frequency receiving part 413, the time and frequency synchronizing means 420 and the space-time-frequency demodulating part 421 are located at the receiving end. The radio frequency receiving part 413 comprises a plurality of groups of receiving antennas (430, 431, 432), receiving local oscillators (417, 418, 419) and combinations of radio frequency processing two (414, 415, 416), wherein each group of combinations corresponds to one OFDM demodulator, each group of combinations receives radio frequency signals through the receiving antennas, and carries out radio frequency processing by the radio frequency processing two according to frequency information of the receiving local oscillators to generate time domain output sequences of the received signals. The time and frequency synchronization device 420 is used for performing time and frequency synchronization, constructing a synchronization objective function with a time point as an argument between each pair of antennas according to the correlation between the time domain output sequence and the local sequence, and estimating a time synchronization point and a frequency offset by using peak information of the synchronization objective function, which will be described in detail later with reference to fig. 10. The time synchronization point and frequency offset information estimated by the time and frequency synchronization means 420 is transmitted to the receiving local oscillator (417, 418, 419) and the OFDM demodulator (422, 423, 424). And the OFDM demodulator combines the estimated time synchronization point and the frequency offset information to carry out OFDM demodulation on the time domain output sequence of the received signal to obtain the restored data information of each antenna, wherein the information of the time synchronization point is mainly used. The mimo decoder 425 performs pilot-assisted channel estimation on the recovered data information of each antenna by using the frequency domain pilot frequency therein, and then performs space-time-frequency decoding to obtain the recovered original input information.

The operation and principle of the partial device in fig. 4 will be described in further detail below.

At the transmitting end, MIMO encoder 402 is operative to generate data information for each antenna based on the frequency domain pilot pattern. Fig. 5 is a schematic diagram of an embodiment of a frequency domain pilot pattern jointly designed for each antenna in a preferred embodiment of the present invention, in which the horizontal axis represents each antenna and the vertical axis corresponds to each subcarrier. As can be seen from fig. 5, each antenna adopts a Pilot-Assisted OFDM (Pilot-Assisted OFDM) transmission structure, and selects subcarrier bits arranged at equal intervalsThe frequency domain pilot frequency (the sub-carrier is called as pilot sub-carrier) is arranged, the positions of the pilot sub-carrier of each antenna are different, and the data information is loaded at the position outside the frequency band containing the pilot sub-carrier. If the number of sub-carriers of OFDM is N, the pilot sub-carrier is N_pThe data position of each antenna is (N-N)_TN_p) And (4) sub-carriers. The interval between pilot subcarriers of each antenna is D_fIs obviously D_fShould be at least greater than the total number of antennas.

The following describes the structure and mapping method of the pilot and data in the frequency domain pilot pattern. The construction of the pilot sequence is first described. Ith transmitting antenna, i ═ 1, 2, K n_TThe method for constructing the pilot sequence included in one transmitted OFDM symbol is as follows: selecting a plurality of sequences with better correlation as basic sequences; repeatedly arranging the selected multiple sections of basic sequences, and multiplying each section of basic sequence by a corresponding coefficient to form a time domain pilot frequency sequence; and performing discrete Fourier transform on the time domain pilot sequence to obtain a frequency domain pilot sequence.

The purpose of this construction method is to make the frequency domain pilot satisfy the predetermined pilot pattern and have good correlation, and the construction principle and details of the frequency domain pilot sequence are described below.

Starting from a frequency domain, the point number of DFT is set as N, the interval of pilot frequency sub-carriers of each antenna is set as M, and the M is required to divide N by N. In order to calculate DFT by using FFT (Fast Fourier Transform), the number of DFT points N is generally required to be 2ⁿTherefore, the condition of dividing M by N is easily satisfied. Frequency domain pilot frequency X of ith antenna_iCan be expressed in the form of an impulse string

Wherein p is_iIndicating a shift of the position of the first pilot subcarrier of the ith transmit antenna with respect to k-0, p_i＝0，1，K M-1，a_r ⁱDenotes an r-th pilot symbol of an ith transmit antenna. The pilots of different antennas are staggered so that p_iMust be different, so this pilot configuration method requires n_T≤M。

X_i(k) The corresponding N-point time domain sequence after IDFT is as follows:

the time domain sequence can be seen to have the following quasi-periodic characteristics: for p_iWhen the time domain sequence is equal to 0, the time domain sequence obviously takes N/M as a period; for p _i1, K, M-1, the time domain sequence is not periodic, but the whole sequence can be regarded as repeating the first N/M points of the sequence M times, each segment is multiplied by

When p is_iThe same holds true for 0, in which case equation (18) is α ═ 1]。

And the corresponding N-point sequence after the frequency domain pilot sequence passes through the IDFT is called a time domain pilot sequence. Both are frequency and time domain representations of the same sequence of mathematical entities.

According to the time domain characteristic of the frequency domain pilot sequence, the N-point pilot sequence can be directly generated in the time domain, so that the PN sequence with better correlation characteristic can be selected to improve the time and frequency synchronization performance. The step of directly generating the pilot frequency sequence in the time domain comprises the following steps:

each transmitting antenna time domain sequence generator generates PN sequence c with length of K-N/M point_i，i＝1，K，n_T. I.e. the base sequence, requires the PN sequence c of each antenna_iThe cross-correlation property between them is good. C is to_iRepeating M times to obtain M sections of same PN sequences, each section is multiplied by corresponding alpha_t ⁱT is 0, K, M-1, thereby forming an N-point time domain pilot sequence x of the antenna i_i(n)。

Generally, as a need for separation and design, a cyclic prefix is added before the pilot sequence, and fig. 6 is a schematic diagram of a time domain pilot sequence with a cyclic prefix in a preferred embodiment of the present invention. In the figure c_qα_t ^qAnd a t-th segment of the time domain pilot sequence of the q-th transmitting antenna. Wherein alpha is_t ⁱIs defined as

Substituting these parameters into equation (17) constitutes x_i(n) is apparent in DFT form X_i(k) In the form of an impulse string distributed at equal intervals, the different antennas are X when used as pilots_i(k) Are staggered with respect to each other.

Next, a mapping method of data in the frequency domain pilot pattern is described: for a certain antenna, the number of pilot subcarriers N_pN/M, the pilots of all antennas occupy N_TN_pOne subcarrier, so that the data symbols for each antenna are mapped to the remaining N-N_TN_p＝N(1-n_T/M) data subcarriers. So that the pilot and data do not overlap in the frequency domain.

The structure of the MIMO encoder 402 in fig. 4 will be described in detail below with reference to pilot sequence generation, setting of predetermined pilot patterns, and a data mapping method. The functional block diagram of the MIMO encoder 402 is shown in fig. 7, which includes a modulator 726, a serial-to-parallel conversion unit 727 and a pilot mapping part.

The modulator 726 performs carrier modulation, such as BPSK (Binary phase shift Keying) modulation, on the original input information to generate a modulated data stream; the serial-parallel conversion device 727 converts the data stream after the carrier modulation into n in serial-parallel mode_TStrip tributary data stream, n_TIs the number of transmit antennas; the pilot frequency mapping part constructs frequency domain pilot frequency sequences of all transmitting antennas, and maps branch data streams of all the transmitting antennas and the frequency domain pilot frequency sequences to corresponding data positions of a preset frequency domain pilot frequency pattern together to generate data information of all the transmitting antennas. The pilot mapping section operates in such a manner that each tributary data stream is deserialized again by

deserializers

728, 729 to N-N_TN_pAnd mapped to corresponding data positions of the frequency domain pilot pattern by the mapping means 732, 733, it will be apparent that the data will occupy N-N of the sub-carriers_TN_pA data location; the pilot sequence generating means 730, 731 generates a time domain pilot sequence corresponding to each antenna, and transforms the time domain pilot sequence into a frequency domain pilot sequence through the DFT means 734, 735, and the frequency domain pilot sequence of each antenna will occupy N_pA data location. It is noted that the pilot sequence generating means may comprise suitable circuitry, logic or processor coupled with or executing suitable programming to generate the time domain pilot sequence.After obtaining the frequency domain pilot sequence, the frequency domain pilot sequence and the mapping data are synthesized by the synthesizing

devices

736 and 737, and data information of each transmitting antenna is generated.

It should be noted that, in fig. 6, only the devices corresponding to two antennas are shown as an example, but such a representation should not be construed as limiting the present invention. The number of transmitting antennas, n, is shown in FIG. 6 by the ellipsis between the two antennas_TAnd may be any natural number. It is obvious that a single transmitting antenna will also fall within the scope of the present invention as a special case.

After generating the data information for each transmit antenna, a plurality of OFDM modulators (403, 404, 405) OFDM modulate the data information for each antenna to generate OFDM symbols, where the OFDM modulation may be implemented by IDFT. After the data mapping is completed, one preferred way is to add a prefix to generate an OFDM symbol. The ith OFDM symbol transmitted by the ith transmitting antenna can be regarded as the sum of two parts of pilot frequency and data, and N is added_gThe transmitted signal after the point cyclic prefix can be expressed as

s_{i}^{l} (n) = d_{i}^{l} (n) + x_{i}^{l} (n), n = 0, K, N + N_{g} - 1, i = K, n_{T} - - - (20)

Wherein d is_l ⁱ(n) and x_l ⁱAnd (n) respectively represents the nth sampling value of the time domain of the data sequence and the pilot sequence contained in the ith OFDM symbol of the ith transmitting antenna.

A schematic block diagram of an OFDM modulator is shown in fig. 8. Since OFDM modulation can be achieved by IDFT, the OFDM modulator includes an IDFT means and a cyclic prefix adding means.

The OFDM symbols generated by modulation are sent to a radio frequency transmitting part, the transmitting part comprises a plurality of groups of combinations of radio frequency processing one (407, 408, 409) and transmitting antennas (410, 411, 412), each group of combinations (such as the combination of the radio frequency processing one 407 and the transmitting antennas 410) corresponds to an OFDM modulator, and the OFDM symbols are subjected to radio frequency processing by the radio frequency processing one and transmitted through the transmitting antennas.

Wherein, the first rf processing is an rf transmitter, and fig. 9 is a schematic block diagram of the rf transmitter according to a preferred embodiment of the present invention. The device comprises a transmitting local oscillator, a first filter and a first amplifier. The input signal and the transmitting local oscillator are synthesized, filtered by the first filter and amplified by the first amplifier to generate a transmitting signal.

The transmitted signal must travel some distance through space before reaching the receiving end. A complex channel space is formed between a signal transmitting end and a signal receiving end, and modeling of the channel space is the basis of the fact that a received signal and a local sequence jointly form a synchronous function. For example, the channel between transmit antenna i to receive antenna j may be a quasi-static (quadrature-static) frequency selective incoherent scattering rayleigh fading channel, i.e., the channel varies randomly at the beginning of each OFDM symbol but remains unchanged for one OFDM symbol time. If L channel impulse responses with different time delays exist in the channel, a tapped delay line model is adopted, and the channel impulse response can be expressed as:

wherein h is_j，i ^l(τ) represents the impulse response of the multipath channel between the transmitting antenna i to the receiving antenna j in the ith OFDM symbol time, L is the number of multipath channels,

represents the equivalent low-pass impulse response of the mth path of the multi-path channel from the transmitting antenna i to the receiving antenna j in the ith OFDM symbol time, t_mIndicating the time delay of the mth diameter.

The transmitted signal may be represented as one (n)_T(N+N_g) X 1 column vector S)_l

S_{l} = {({(\begin{matrix} s_{1}^{l} (0) \\ M \\ S_{n_{T}}^{l} \end{matrix})}^{T}, {(\begin{matrix} s_{1}^{l} (1) \\ M \\ s_{n_{T}}^{l} (1) \end{matrix})}^{T}, K, {(\begin{matrix} s_{1}^{l} (N + N_{g} - 1) \\ M \\ s_{n_{T}}^{l} (N + N_{g} - 1) \end{matrix})}^{T})}^{T} - - - (22)

Wherein, (g)^TRepresenting a transposition operation, s_i ^lAnd (n) represents the nth sampling value of the time domain of the ith OFDM symbol sent by the ith transmitting antenna at the ith OFDM symbol time.

The channel impulse response model shown in equation (21) can be expressed in the form of a time-varying FIR (Finite impulse response) filter, thus n_TStrip transmitting antenna n_RThe mimo channel model of the strip receive antenna can be expressed as (n)_R(N+N_g+N_L-1))×(n_T(N+N_g) H) of a matrix_lIn which N is_LThe number of delay sample points representing the lth path,

H_{l} = (\begin{matrix} H (0) & 0 & L & 0 \\ H (1) & H (0) & L & M \\ M & H (1) & L & 0 \\ H (N_{L} - 1) & M & L & H (0) \\ 0 & H (N_{L} - 1) & L & H (1) \\ M & M & L M \\ 0 & 0 & L & H (N_{L} - 1) \end{matrix}) - - - (23)

wherein,

H (m) = (\begin{matrix} h_{1,1}^{l} (m) & K & h & _{{1, n}_{T}}^{l} (m) \\ M & O & M \\ h_{n_{R}, 1}^{l} (m) & L & h_{n_{R}, n_{T}}^{l} (m) \end{matrix}), m = 0, . . ., N_{L} - 1 - - - (24)

considering additive complex white gaussian noise, the received signal R is perfectly synchronized in time and frequency_lCan be expressed as (n)_R(N+N_g+N_L-1)) × 1 column vector

R_l＝H_lS_l+W_l (25)

Wherein, W_lThe elements of (1) are zero-mean additive complex Gaussian white noise samples with variance of σ_w ²Can be expressed as

W_{l} = {({(\begin{matrix} w_{1}^{l} (0) \\ M \\ w_{n_{R}}^{l} (0) \end{matrix}),}^{T} {(\begin{matrix} w_{1}^{l} (1) \\ M \\ w_{n_{R}}^{l} (1) \end{matrix})}^{T}, K, {(\begin{matrix} w_{1}^{l} (N + N_{g} + N_{L} - 2) \\ M \\ w_{n_{R}}^{l} (N + N_{g} + N_{L} - 2) \end{matrix})}^{T})}^{T} - - - (26)

After being propagated through a channel, the signal reaches a receiving end from a transmitting end. At the receiving end, the electromagnetic wave is first received by the radio frequency receiving portion 413, generating a time domain output sequence. The radio frequency receiving part 413 comprises a plurality of groups of receiving antennas (430, 431, 432), receiving local oscillators (417, 418, 419) and combinations of radio frequency processing two (414, 415, 416), wherein each group of combinations corresponds to one OFDM demodulator, each group of combinations receives radio frequency signals through the receiving antennas, and carries out radio frequency processing by the radio frequency processing two according to frequency information of the receiving local oscillators to generate time domain output sequences of the received signals.

And the second radio frequency processing is a radio frequency receiver. Fig. 11 is a schematic block diagram of a radio frequency receiver in a preferred embodiment of the invention. The input signal is firstly amplified by the amplifier and then filtered by the second filter, the filtered signal is combined with the signal from the receiving local oscillator, and the synthesized signal is filtered by the second filter to obtain the output of the radio frequency receiver.

Then, the time and frequency synchronization device 420 performs time and frequency synchronization, constructs a synchronization objective function with a time point as an argument between each pair of antennas according to the correlation between the time domain output sequence and the reference sequence, and estimates a time synchronization point and a frequency offset using peak information of the synchronization objective function. The operation of the time and frequency synchronization device 420 will be described in detail below.

When the time and frequency are completely synchronized, the received signal of the jth receiving antenna at the ith symbol time can be expressed as shown in equations (22) to (26)

Since l and n are both time variables, there are

r_{j}^{l} (n) = r_{j} (n + l (N + N_{g}))

. For convenience, let N ═ N + l (N + N)_g) Then n' is the number of the actual time domain sampling point of the OFDM symbol: n' is 0, 1, K.

Considering time delay and frequency offset, let the receiving delay of the jth receiving antenna to the ith transmitting antenna be theta_j，iSetting the normalized frequency deviation of the radio frequency receiver as epsilon_j，i，i＝1，...，n_T，j＝1，...，n_R. The received signal of the jth receiving antenna can be expressed as

Wherein * g * indicates rounding down.

The following describes a specific algorithm of synchronization by taking synchronization between the jth receiving antenna and the qth transmitting antenna as an example, and the method between other pairs of antennas is similar to the above. From the formula (28), it can be found

<math> <mrow> <mo>+</mo> <munderover> <mi>Σ</mi> <munder> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <mrow> <mi>i</mi> <mo>&NotEqual;</mo> <mi>q</mi> </mrow> </munder> <msub> <mi>n</mi> <mi>T</mi> </msub> </munderover> <munderover> <mi>Σ</mi> <mrow> <mi>m</mi> <mo>=</mo> <mn>0</mn> </mrow> <mrow> <msub> <mi>N</mi> <mi>L</mi> </msub> <mo>-</mo> <mn>1</mn> </mrow> </munderover> <msubsup> <mi>h</mi> <mrow> <mi>j</mi> <mo>,</mo> <mi>i</mi> </mrow> <mi>l</mi> </msubsup> <mrow> <mo>(</mo> <mi>m</mi> <mo>)</mo> </mrow> <msub> <mi>s</mi> <mi>i</mi> </msub> <mrow> <mo>(</mo> <msup> <mi>n</mi> <mo>′</mo> </msup> <mo>-</mo> <mi>m</mi> <mo>-</mo> <msub> <mi>θ</mi> <mrow> <mi>j</mi> <mo>,</mo> <mi>i</mi> </mrow> </msub> <mo>)</mo> </mrow> <msup> <mi>e</mi> <mrow> <mi>j</mi> <mn>2</mn> <msub> <mi>πϵ</mi> <mrow> <mi>j</mi> <mo>,</mo> <mi>i</mi> </mrow> </msub> <msup> <mi>n</mi> <mo>′</mo> </msup> <mo>/</mo> <mi>N</mi> </mrow> </msup> <mo>+</mo> <msub> <mi>w</mi> <mi>j</mi> </msub> <mrow> <mo>(</mo> <msup> <mi>n</mi> <mo>′</mo> </msup> <mo>)</mo> </mrow> </mrow> </math>

<math> <mrow> <mo>+</mo> <munderover> <mi>Σ</mi> <munder> <mrow> <mi>m</mi> <mo>=</mo> <mn>0</mn> </mrow> <mrow> <mi>m</mi> <mo>&NotEqual;</mo> <msub> <mi>m</mi> <mi>max</mi> </msub> </mrow> </munder> <mrow> <msub> <mi>N</mi> <mi>L</mi> </msub> <mo>-</mo> <mn>1</mn> </mrow> </munderover> <msubsup> <mi>h</mi> <mrow> <mi>j</mi> <mo>,</mo> <mi>q</mi> </mrow> <mi>l</mi> </msubsup> <mrow> <mo>(</mo> <mi>m</mi> <mo>)</mo> </mrow> <msub> <mi>s</mi> <mi>q</mi> </msub> <mrow> <mo>(</mo> <msup> <mi>n</mi> <mo>′</mo> </msup> <mo>-</mo> <mi>m</mi> <mo>-</mo> <msub> <mi>θ</mi> <mrow> <mi>j</mi> <mo>,</mo> <mi>q</mi> </mrow> </msub> <mo>)</mo> </mrow> <msup> <mi>e</mi> <mrow> <mi>j</mi> <mn>2</mn> <msub> <mi>πϵ</mi> <mrow> <mi>j</mi> <mo>,</mo> <mi>q</mi> </mrow> </msub> <msup> <mi>n</mi> <mo>′</mo> </msup> <mo>/</mo> <mi>N</mi> </mrow> </msup> </mrow> </math>

<math> <mrow> <mo>+</mo> <munderover> <mi>Σ</mi> <munder> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <mrow> <mi>i</mi> <mo>&NotEqual;</mo> <mi>q</mi> </mrow> </munder> <msub> <mi>n</mi> <mi>T</mi> </msub> </munderover> <munderover> <mi>Σ</mi> <mrow> <mi>m</mi> <mo>=</mo> <mn>0</mn> </mrow> <msub> <mi>N</mi> <mi>L</mi> </msub> </munderover> <msubsup> <mi>h</mi> <mrow> <mi>j</mi> <mo>,</mo> <mi>i</mi> </mrow> <mi>l</mi> </msubsup> <mrow> <mo>(</mo> <mi>m</mi> <mo>)</mo> </mrow> <msub> <mi>s</mi> <mi>i</mi> </msub> <mrow> <mo>(</mo> <msup> <mi>n</mi> <mo>′</mo> </msup> <mo>-</mo> <mi>m</mi> <mo>-</mo> <msub> <mi>θ</mi> <mrow> <mi>j</mi> <mo>,</mo> <mi>i</mi> </mrow> </msub> <mo>)</mo> </mrow> <msup> <mi>e</mi> <mrow> <mi>j</mi> <mn>2</mn> <msub> <mi>πϵ</mi> <mrow> <mi>j</mi> <mo>,</mo> <mi>i</mi> </mrow> </msub> <msup> <mi>n</mi> <mo>′</mo> </msup> <mo>/</mo> <mi>N</mi> </mrow> </msup> <mo>+</mo> <msub> <mi>w</mi> <mi>j</mi> </msub> <mrow> <mo>(</mo> <msup> <mi>n</mi> <mo>′</mo> </msup> <mo>)</mo> </mrow> </mrow> </math>

Wherein m is_maxIs the strongest path delay of the multipath channel between the transmitting antenna q and the receiving antenna j at the ith OFDM symbol time. The strongest path is defined as

m_{\max} = \arg \max_{m} {E [{| h_{j, q}^{l} (m) |}^{2}]}

，E[g]Representing a mathematical expectation. Substituting equation (20) into equation (29), r_j(n') may be further represented as

<math> <mrow> <mo>+</mo> <msub> <mi>I</mi> <mi>Data</mi> </msub> <mo>+</mo> <msub> <mi>I</mi> <mi>ISI</mi> </msub> <mo>+</mo> <msub> <mi>I</mi> <mi>OtherAntenna</mi> </msub> <mo>+</mo> <msub> <mi>w</mi> <mi>j</mi> </msub> <mrow> <mo>(</mo> <msup> <mi>n</mi> <mo>′</mo> </msup> <mo>)</mo> </mrow> </mrow> </math>

Wherein,

<math> <mrow> <msub> <mi>I</mi> <mi>ISI</mi> </msub> <mo>=</mo> <munderover> <mi>Σ</mi> <munder> <mrow> <mi>m</mi> <mo>=</mo> <mn>0</mn> </mrow> <mrow> <mi>m</mi> <mo>&NotEqual;</mo> <msub> <mi>m</mi> <mi>max</mi> </msub> </mrow> </munder> <mrow> <msub> <mi>N</mi> <mi>L</mi> </msub> <mo>-</mo> <mn>1</mn> </mrow> </munderover> <msubsup> <mi>h</mi> <mrow> <mi>j</mi> <mo>,</mo> <mi>q</mi> </mrow> <mi>l</mi> </msubsup> <mrow> <mo>(</mo> <mi>m</mi> <mo>)</mo> </mrow> <msub> <mi>s</mi> <mi>q</mi> </msub> <mrow> <mo>(</mo> <msup> <mi>n</mi> <mo>′</mo> </msup> <mo>-</mo> <mi>m</mi> <mo>-</mo> <msub> <mi>θ</mi> <mrow> <mi>j</mi> <mo>,</mo> <mi>q</mi> </mrow> </msub> <mo>)</mo> </mrow> <msup> <mi>e</mi> <mrow> <mi>j</mi> <mn>2</mn> <msub> <mi>πϵ</mi> <mrow> <mi>j</mi> <mo>,</mo> <mi>q</mi> </mrow> </msub> <msup> <mi>n</mi> <mo>′</mo> </msup> <mo>/</mo> <mi>N</mi> </mrow> </msup> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>31</mn> <mo>)</mo> </mrow> </mrow> </math>

<math> <mrow> <msub> <mi>I</mi> <mi>OtherAntenna</mi> </msub> <mo>=</mo> <munderover> <mi>Σ</mi> <munder> <mrow> <mi>i</mi> <mo>=</mo> <mn>0</mn> </mrow> <mrow> <mi>i</mi> <mo>&NotEqual;</mo> <mi>q</mi> </mrow> </munder> <msub> <mi>n</mi> <mi>T</mi> </msub> </munderover> <munderover> <mi>Σ</mi> <mrow> <mi>m</mi> <mo>=</mo> <mn>0</mn> </mrow> <mrow> <msub> <mi>N</mi> <mi>L</mi> </msub> <mo>-</mo> <mn>1</mn> </mrow> </munderover> <msubsup> <mi>h</mi> <mrow> <mi>j</mi> <mo>,</mo> <mi>i</mi> </mrow> <mi>l</mi> </msubsup> <mrow> <mo>(</mo> <mi>m</mi> <mo>)</mo> </mrow> <msub> <mi>s</mi> <mi>i</mi> </msub> <mrow> <mo>(</mo> <msup> <mi>n</mi> <mo>′</mo> </msup> <mo>-</mo> <mi>m</mi> <mo>-</mo> <msub> <mi>θ</mi> <mrow> <mi>j</mi> <mo>,</mo> <mi>i</mi> </mrow> </msub> <mo>)</mo> </mrow> <msup> <mi>e</mi> <mrow> <mi>j</mi> <mn>2</mn> <msub> <mi>πϵ</mi> <mrow> <mi>j</mi> <mo>,</mo> <mi>q</mi> </mrow> </msub> <msup> <mi>n</mi> <mo>′</mo> </msup> <mo>/</mo> <mi>N</mi> </mrow> </msup> </mrow> </math>

I_Datarepresenting the interference of the data of the strongest path of the q-th transmitting antenna on the pilot signal, I_ISIAnd representing the interference of other paths of the q-th transmitting antenna to the strongest path signal. I is_{Other Antenna}Representing the interference of the transmission signals of other transmission antennas to the q-th transmission antenna. Order to

<math> <mrow> <msubsup> <mi>w</mi> <mi>j</mi> <mo>%</mo> </msubsup> <mrow> <mo>(</mo> <msup> <mi>n</mi> <mo>′</mo> </msup> <mo>)</mo> </mrow> <mo>=</mo> <msub> <mi>I</mi> <mi>Data</mi> </msub> <mo>+</mo> <msub> <mi>I</mi> <mi>ISI</mi> </msub> <mo>+</mo> <msub> <mi>I</mi> <mi>OtherAntenna</mi> </msub> <mo>+</mo> <msub> <mi>w</mi> <mi>j</mi> </msub> <mrow> <mo>(</mo> <msup> <mi>n</mi> <mo>′</mo> </msup> <mo>)</mo> </mrow> </mrow> </math>

Then r is_j(n') may be represented as

According to the above with reference to fig. 6Description of the Pilot sequences, x_q ^l(n) the part excluding the cyclic prefix has a quasi-periodic nature, c_qα_t ^qAnd t is 0, K, M-1 in the t section of the time domain pilot sequence of the q antenna. Wherein c is_qIs the PN sequence of the qth transmitting antenna, and the length is N/M. Let K equal N/M, then x_q ^l(n) can be represented as

Setting a sliding window with the length of N for the qth transmitting antenna, and constructing the sliding window continuously changing along with the time variable N':

let eta be_j，q(n′)＝|λ_j，q(n') |, then η_j，qAnd (n') is the synchronization objective function.

With the synchronization objective function, the time synchronization point and the frequency offset can be estimated by using the peak information of the synchronization objective function. A method for judging the correlation peak of synchronous target function directly sets a threshold, when the target function value exceeds the threshold, the target function value is judged to be synchronous, and the corresponding sampling point is a time synchronous point.

The threshold is generated by self-adaptive threshold mode, and the method is to synchronize the current output value of the objective function with the previous N₁T of point mean_hAnd (4) comparing times, judging that the time synchronization is achieved when the time synchronization is larger than or equal to the preset time, and judging the conditions as follows:

<math> <mrow> <mo>|</mo> <msub> <mi>λ</mi> <mrow> <mi>j</mi> <mo>,</mo> <mi>q</mi> </mrow> </msub> <mrow> <mo>(</mo> <mo></mo> <msup> <mi>n</mi> <mo>′</mo> </msup> <mo></mo> <mo>)</mo> </mrow> <mo>|</mo> <mo>&GreaterEqual;</mo> <msub> <mi>T</mi> <mi>h</mi> </msub> <mfrac> <mrow> <munderover> <mi>Σ</mi> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <msub> <mi>N</mi> <mn>1</mn> </msub> </munderover> <mo>|</mo> <msub> <mi>λ</mi> <mrow> <mi>j</mi> <mo>,</mo> <mi>q</mi> </mrow> </msub> <mrow> <mo>(</mo> <msup> <mi>n</mi> <mo>′</mo> </msup> <mo>-</mo> <mi>i</mi> <mo>)</mo> </mrow> <mo>|</mo> </mrow> <msub> <mi>N</mi> <mn>1</mn> </msub> </mfrac> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>35</mn> <mo>)</mo> </mrow> </mrow> </math>

when the time synchronization is accurately completed, the objective function eta_j，q(n') the maximum peak should occur. For the first OFDM symbol, the peak should be m ═ n ═ m_max+θ_j，q+N_gAnd occurs. From equation (32) equation (34),

at this time

Due to the fact that

By using the formula (33), the above formula can be simplified to

Wherein, w_j' is the sum of all additive independent terms. It can be seen that formula (37) has a peak output at the time of fine synchronization, and according to the spread spectrum principle, the first term of formula (37) is much larger modulo than the second term, so that the interference term is ignored, and then at the time of fine time synchronization, the synchronization objective function value is:

it can be seen that the objective function is related to the normalized frequency offset epsilon and the strongest path fading. For example, if N1024 and K128, the first zero point of the target function that varies with e is obtained at 8. In the new generation mobile communication, the carrier frequency is usually 3 to 3.5GHz, and if the carrier frequency is 3.5GHz, the stability factor of the receiver crystal oscillator is 10ppm (parts per million), and the OFDM subcarrier spacing is 20KHz, then ∈ is 1.75. At this time, as long as the strongest path does not generate deep fading, the peak value of the time synchronization target function is still very high, and the requirement of time synchronization is met.

Since the received signal has quasi-periodicity, the objective function η is seen by formula (32), formula (33) and formula (34)_j，q(n') has multiple peaks, which requires that the threshold setting needs to be quite accurate, otherwise, a simple threshold determination method is easy to synchronize to a side lobe. In order to obtain an accurate synchronization point, the method needs to be improved, a search window is set, and the improved method comprises the following steps: first, n satisfying formula (35) is found₁' Point; in the interval

Chinese search η_j，q(n') maximum point n₂'; point n₂' i.e. the corresponding synchronization point (synchronized to the strongest path).

This improved method achieves synchronization to the strongest path, but the strongest path is not necessarily the first path, although the strongest path is typically the first path. Due to the cyclic prefix, demodulation of OFDM does not need to be strictly synchronized to the first path. However, if the first path cannot be synchronized, the multipath energy before the strongest path cannot be utilized, and interference is generated on signal demodulation. To this end, in a preferred embodiment of the present invention, the following method for first path synchronization is provided: firstly, finding the strongest path; from most intense path to N₂Searching points backwards; setting and judging the ratio of the boiling to the pre-boiling by taking the boiling as an objective function₃N of the mean of the points₄The size is doubled; find out to satisfyAnd the point of the judgment condition is the first path, otherwise, the strongest path is the first path.

This is because, even if there is a multipath component in front of the strongest path, the objective function value corresponding to the gain does not satisfy the determination condition, indicating that the gain is much smaller than the strongest path and can be ignored.

It is assumed that the time synchronization of the antennas is completed. As can be seen from (37), epsilon_j，qThe estimation can be performed as follows:

wherein arg (g) is the phase angle calculation.

In accordance with the above-mentioned time and frequency synchronization method, a block diagram of an embodiment of a synchronization algorithm of the time and frequency synchronization apparatus 420 according to a preferred embodiment of the present invention is shown in fig. 10. Wherein the module 1038, the module 1039, the module 1040, and the module 1041 are used for constructing λ in the formula (34)_j，q(n'), Module 1042 vs. λ_j，q(n') taking the absolute value to obtain the synchronous objective function | lambda_i(k) L. Thereafter, a block 1043 is provided for_i(k) I calculate the decision threshold value, then compareAnd the comparison module judges according to the formula (35), finds the peak value of the synchronous target function and preliminarily determines the synchronous point. After the synchronization point is preliminarily determined, a first path between the antennas is further found through the module 1045. Meanwhile, the module 1046 obtains a phase angle of the peak value of the synchronous objective function to obtain a frequency offset estimation value.

The synchronization method of the embodiment of the invention can simultaneously realize the frequency offset estimation of the integer part and the frequency offset estimation of the decimal part. The frequency offset estimation range is

If N is 1024 and K is 128, then

. The carrier frequency is 3.5GHz, the stability coefficient of the crystal oscillator of the receiver is 10ppm (parts per million), the OFDM subcarrier interval is 20KHz, and epsilon is 1.75 < 4, which shows that the frequency synchronization of the invention can meet the frequency synchronization requirement of the new generation of mobile communication.

In the process of time and frequency synchronization, the accuracy of the frequency synchronization algorithm can be improved by using a method of transmitting and receiving diversity. For MIMO systems with the same frequency offset between each pair of transmit and receive antennas, there are

After the first frequency synchronization, residual frequency offset exists due to the influence of interference, which can be realized by secondary frequency synchronization. Assuming that the first frequency synchronization can completely offset the integer part of the frequency offset, after the receiving end compensates the frequency offset, the residual frequency offset can be achieved by performing frequency offset estimation again by using the far data. Rewrite equation (34) to

Similar to the first frequency offset estimation, may be derived

The calculation formula (c) is as follows:

where P is used to adjust the accuracy. The larger the P, the higher the accuracy, but the smaller the frequency offset estimation range. Thus, the

Can be expressed as the sum of two parts:

and

respectively, representing the results of the first and second frequency offset estimations.

The method for constructing the objective function is to construct the objective function for each pair of transmitting and receiving antennas respectively, wherein the time synchronization of each transmitting antenna is realized based on the suppression of other antenna data. When the number of transmitting antennas is large, the interference increases and the performance is degraded, and as shown in equation (38), the peak value of the objective function can be made larger by increasing K. The synchronization accuracy is also improved by reconstructing the synchronization function using the diversity gain as follows. The specific conditions are specifically described below.

For example, when the delay from the same transmit antenna to each receive antenna is the same, receive diversity may be utilized, and the synchronization objective function of the qth transmit antenna is:

similarly, when the delay from each transmitting antenna to the jth receiving antenna is the same, transmit diversity may be utilized, and the synchronization objective function of the jth receiving antenna at this time is:

when the time delay from each transmitting antenna to each receiving antenna is the same and the frequency offset between each pair of transceiving antennas is the same, the transmit diversity can be further utilized, and the synchronization objective function is:

in particular, for the MIMO system with the same frequency offset between each pair of transmit-receive antennas, if N is even, another method for frequency synchronization is available:

it can be seen from equation (18) that the N-point time domain pilot sequence has the following properties: when p is_iWhen the number of the pilot frequency sequences is odd, the front N/2 point of the time domain pilot frequency sequence is equal to the rear N/2 point multiplied by-1; when p is_iAnd when the number is even, the front N/2 point and the rear N/2 point of the time domain pilot frequency sequence are the same.

The number of transmitting antennas is limited by the spacing M of the pilot subcarriers, theoretically transmittingNumber of antennas n_TM is less than or equal to, when n is_TWhen M is no more data subcarriers, at least n is required to be satisfied in order not to reduce the spectrum utilization_T≤M/2。

The position of the first pilot subcarrier of the transmitting antenna in the predetermined pilot pattern is then shifted with respect to the origin of the OFDM subcarrier, i.e. the position p of the pilot subcarrier of the ith antenna_iSet of (i) ═ 0.. times.n_T-1, which can be optionally selected from one of the following two groups:

p_ie {0, 2.., M-2} or p_i∈{1，3，...，M-1} (46)

Namely: p is a radical of_iMust be either all odd or all even.

Let the time synchronization point be n₁If the time synchronization of each antenna is completed, the frequency offset calculation formula is:

wherein,

equation (46) can be further written as

The method fully utilizes diversity gain, the more the number of transmitting antennas is, the more accurate the synchronization is, and the frequency deviation estimation range is

After obtaining the time synchronization point and frequency offset information, the time synchronization point and frequency offset information estimated by the time and frequency synchronization apparatus 420 are transmitted to the receiving local oscillator (417, 418, 419) and the OFDM demodulator (422, 423, 424). And the OFDM demodulator combines the estimated time synchronization point and the frequency offset information to carry out OFDM demodulation on the time domain output sequence of the received signal to obtain the restored data information of each antenna. Then, the mimo decoder 425 performs pilot-assisted channel estimation on the restored data information of each antenna by using the frequency domain pilot frequency therein, and then performs space-time-frequency decoding to obtain the restored original input information.

In a preferred embodiment of the present invention, a schematic block diagram of an OFDM demodulator is shown in fig. 12. The OFDM demodulator of fig. 12 includes a cyclic prefix removal module and a DFT module, corresponding to the OFDM modulator of fig. 8. The input signal is first stripped of the cyclic prefix and then subjected to a DFT to produce an output signal.

It can be seen from the foregoing embodiments that, in the present invention, by designing the special pilot sequences of each transmitting antenna, the system can implement time and frequency synchronization with lower overhead, and each antenna only needs to send an OFDM symbol with pilot inserted in the frequency domain, and does not need additional training sequences, and can complete the MIMO channel estimation and the transmission of the bearer data service while completing the synchronization task. The time and frequency synchronization method is suitable for a complex actual MIMO system, the time delay from each transmitting antenna to each receiving antenna can be different, the frequency deviation between each pair of transmitting and receiving antennas can also be different, and the diversity gain of the antennas can be utilized to further improve the synchronization performance when the frequency deviation or the time delay is the same. Compared with the traditional synchronization method which only can capture the strongest path, the embodiment of the invention provides a method for capturing the first path. Meanwhile, compared with the existing method, the method has the advantages of being suitable for multipath channels and not needing to estimate the signal-to-noise ratio.

It should be noted that, in the process of describing the embodiment of the present invention with reference to fig. 4, as an example, the radio frequency processing one, the radio frequency processing two, the OFDM modulator, the OFDM demodulator, the local oscillator, and other devices in each link only represent a certain number (for example, only 2 transmitters 602 and 603 are labeled), but such labels should not be construed as limiting the scope of the present invention. The scope of the present invention is intended to include all combinations of the number of transmitting antennas and receiving antennas that satisfy the conditions of the present invention. In particular, the case of a single input antenna or a single output antenna is taken as a particular example of the present invention, to which the methods and systems relating to the various aspects of the present invention are fully applicable and which therefore also fall within the scope of protection of the present invention.

Claims

1. A method for time and frequency synchronization in orthogonal frequency division multiplexing communications, characterized by:

estimating a time synchronization point and a frequency offset by using peak value information of a synchronization objective function;

2. The method of claim 1, wherein the step of constructing the frequency-domain pilot sequences for the transmit antennas comprises:

selecting a plurality of sequences as a basic sequence;

repeatedly arranging the basic sequences, and multiplying each section of basic sequence by a corresponding coefficient to form a time domain pilot frequency sequence;

and performing discrete Fourier transform on the time domain pilot sequence to obtain a frequency domain pilot sequence.

3. The method according to claim 1, wherein in the predetermined frequency domain pilot pattern, the frequency domain pilot of each antenna is located at the position of equally spaced subcarriers, and the subcarriers are called pilot subcarriers; the positions of the pilot frequency sub-carriers of the antennas are different from each other; the data information is located at a position outside the frequency band containing the pilot subcarriers.

4. The method of claim 1, wherein the step of estimating the time synchronization point and the frequency offset using the peak information of the synchronization objective function comprises:

judging a correlation peak of a synchronization target function by using a self-adaptive threshold method, and estimating a time synchronization point according to corresponding peak value information; performing first frequency offset estimation by taking a phase angle from a peak value of a synchronous target function by using the obtained time synchronization point; and performing second frequency offset estimation by using data farther away from the current moment than the first frequency offset estimation by adopting the same method as the first frequency offset estimation.

5. The method of claim 1, wherein the synchronization objective function is constructed using receive diversity if the time delay from the same transmit antenna to each receive antenna is the same; if the time delay from each transmitting antenna to the same receiving antenna is the same, then using transmit diversity; if the time delay and frequency offset are the same between each pair of transmit and receive antennas, full diversity, including receive diversity and transmit diversity, is used.

6. The method of claim 3, wherein when the frequency offsets between the transceiving antennas of each pair are the same and the length of the time domain pilot sequence is even, the predetermined frequency domain pilot pattern is designed such that the displacements of the first pilot subcarrier positions of the respective transmitting antennas relative to the origin of the OFDM subcarriers are all odd numbers or all even numbers; and a synchronization objective function is constructed with full diversity gain.

7. The method of claim 4, wherein the time synchronization point is estimated by determining a sampling point where a peak of the synchronization objective function is located as the time synchronization point.

8. The method of claim 4, wherein the step of estimating a time synchronization point comprises: and setting a search window in a specific range near a sampling point where a peak point of the synchronous target function is located, and searching a maximum point of the synchronous target function in the search window, wherein the maximum point corresponds to a synchronous point which is synchronized to the strongest path between the antennas.

9. The method of claim 8, wherein the specific range is a length of each half of the time domain pilot sequence around the sampling point where the peak of the synchronous objective function is located.

10. The method of claim 8, wherein after finding the strongest path synchronization point between the antennas, further comprising the step of finding a first path between the antennas by:

comparing the synchronization target function with the former N₃N of the mean of the points₄Multiplying the size as a first diameter judgment condition; n before the sampling point corresponding to the strongest path₂The point starts to search backwards until the sampling point corresponding to the strongest path is self; if there is a sampling point satisfying the first path determination condition, the sampling point corresponds to the first path, otherwise the strongest path is the first path, where N is₂、N₃、N₄Are the corresponding coefficients.

11. The method of claim 1, wherein the step of constructing the frequency-domain pilot sequences of the transmitting antennas further comprises: a cyclic prefix is added before the pilot sequence.

12. An orthogonal frequency division multiplexing communication system capable of realizing time and frequency synchronization, characterized in that,

its transmitting end includes:

a multiple-input multiple-output encoder for encoding the original input information and generating data information for each antenna from the frequency domain pilot pattern,

at least one OFDM modulator for OFDM modulating data information for each antenna to generate OFDM symbols,

and a radio frequency transmitting part for performing radio frequency processing on the OFDM symbol and transmitting the OFDM symbol through a transmitting antenna;

its receiving end includes:

a radio frequency receiving portion that receives a radio frequency signal and performs radio frequency processing to generate a time domain output sequence of the received signal,

a time and frequency synchronization device which constructs a synchronization objective function with a time point as an independent variable among each pair of antennas according to the correlation between the time domain output sequence and the local sequence and estimates a time synchronization point and a frequency offset by using peak information of the synchronization objective function,

at least one OFDM demodulator for OFDM demodulating the time domain output sequence of the received signal by combining the estimated time synchronization point and the frequency offset information to obtain the restored data information of each antenna,

and a multi-input multi-output decoder for performing pilot-assisted channel estimation on the restored data information of each antenna by using the frequency domain pilot frequency and then performing space-time-frequency decoding to obtain the restored original input information.

13. The orthogonal frequency division multiplexing communication system capable of achieving time and frequency synchronization according to claim 12, wherein the multiple-input multiple-output encoder comprises:

a modulation module for carrier modulating the original input information to produce a modulated data stream,

original information after carrier modulation is serial-parallel converted into n_TSerial-to-parallel conversion module for a strip-tributary data stream, n of which_TIn order to be able to transmit the number of antennas,

and the pilot mapping module part is used for constructing frequency domain pilot sequences of all the transmitting antennas and mapping the branch data streams of all the transmitting antennas and the frequency domain pilot sequences to corresponding data positions of a preset frequency domain pilot pattern together so as to generate data information of all the transmitting antennas.

14. The orthogonal frequency division multiplexing communication system that can achieve time and frequency synchronization according to claim 12,

each OFDM modulator comprises: an IDFT device for performing inverse discrete Fourier transform on data information of an antenna, and a cyclic prefix adding device for adding a cyclic prefix before a pilot sequence output by the IDFT device;

each OFDM demodulator comprises: a cyclic prefix removing means for removing the cyclic prefix, and a DFT means for performing discrete fourier transform.

15. The orthogonal frequency division multiplexing communication system capable of achieving time and frequency synchronization according to claim 12, wherein the time and frequency synchronization means comprises:

a correlation module for performing correlation processing on the signal inputted from the radio frequency receiving section,

a delay module for delaying an output of the correlation module,

a conjugate extracting module for extracting conjugate from the output of the correlation module,

a multiplication module for performing multiplication operation on the outputs of the delay module and the conjugate taking module,

a summation module for summing the sequence values output by the multiplication module,

a modulus module for performing modulus processing on the output of the summation module,

a threshold module for calculating a threshold according to the output of the modulus module,

a comparison module for comparing the output of the modulus taking module and the threshold module;

a time synchronization module for finding a first path according to the output of the comparison module and further obtaining a time synchronization point;

and the frequency offset estimation module is used for obtaining a phase angle finger according to the output of the comparison module so as to obtain a frequency offset estimation value.