CN1259780C - New OFDM time, frequency synchronization method - Google Patents

Abstract

The present invention provides a new OFDM time and frequency synchronization method. Data on Ng points behind OFDM raw data and a training sequence are weighted and superposed to form guard spacing point to point on a sending terminal of an OFDM system, and OFDM raw data is kept unchanged. A receiving terminal of the OFDM system uses a merging algorithm, and the correlativity of data in the OFDM guard spacing and data of rear Ng points and the correlativity induced by the training sequence are comprehensively used for treating received data so as to realize the synchronization of time and frequency. The method of the present invention enables the OFDM system to have the advantages of good synchronization performance and small interference of the training sequence to the data.

Description

A Method of OFDM Time and Frequency Synchronization

technical field

The invention belongs to the field of wireless communication or wired communication, and particularly relates to the time-frequency synchronization technology in the OFDM system.

Background technique

OFDM has attracted more and more attention due to its high data transmission rate, strong anti-multipath interference ability, and high spectrum efficiency. It has been successfully used in wired and wireless communications. Such as: DAB (Digital Audio Broadcasting), DVB, EEE802.11a and HyperLAN/2, in IEEE802.16 currently being formulated, OFDM technology is also involved in a large number. OFDM, a new modulation technology, can also be used in a new generation of mobile communication systems. The use of OFDM technology will greatly improve the transmission data rate and spectral efficiency of the new generation of mobile communication systems, and has a good ability to resist multipath, co-channel interference and impact noise, see the literature: Bingham, J.AC., "Multicarrier modulation for data transmission: an idea whose time has come," IEEE Communications Magazine, Volume: 28 Issue: 5, May 1990, Page(s): 5-14; and literature: YunHee Kim; Iickho Song; Hong Gil Kim; Taejoo Chang; Hyung Myung Kim, "Performance analysis of a coded OFDM system in time-varying multipath Rayleigh fading channels," Vehicular Technology, IEEE Transactions on, Volume: 48 Issue: 5, Sept, 1999, Page(s): 1610-1615.

OFDM synchronization is divided into time synchronization and frequency synchronization. The location of the synchronization module is shown in module 11 in FIG. 1 . The purpose of time synchronization is to find the boundaries of each OFDM symbol in the received serial data stream; and the purpose of frequency synchronization is to find and correct the frequency offset at the receiving end. One of the weaknesses of OFDM technology is the requirement for time and frequency synchronization, especially frequency synchronization, which is much higher than that of single-carrier systems. It is generally required that the frequency offset of the system using OFDM technology does not exceed 2% of its subcarrier spacing, see the literature van de Beek, J.J.; Sandell, M.; Borjesson, P.O., "ML estimation of time and frequency offset in OFDM systems , "Signal Processing, IEEE Transactions on, Volume: 45 Issue: 7, July 1997, Page(s): 1800-1805. One of the synchronization techniques of OFDM is to use the redundancy of the guard interval for time and frequency synchronization, see the literature van de Beek, J.J.; Sandell, M.; Borjesson, P.O., "ML estimation of time and frequency offset in OFDMsystems," Signal Processing, IEEE Transactions on, Volume: 45 Issue: 7, July 1997, Page(s): 1800-1805.

In OFDM technology, in order to eliminate inter-symbol interference and co-channel interference, a guard interval is generally added before each OFDM symbol. The length of the guard interval is generally required to be greater than 2 or 4 times the length of the channel impulse response. The guard interval content is generally a part of the OFDM symbol.

There are two synchronization methods for conventional OFDM:

1) Using the correlation between the guard interval and OFDM symbols, time and frequency synchronization can be realized. See vande Beek, JJ; Sandell, M.; Borjesson, PO, "ML estimation of time and frequency offset in OFDMsystems," Signal Processing, IEEE Transactions on, Volume: 45 Issue: 7, July 1997, Page(s): 1800-1805. Place guard intervals as in Figure 2. Since the data 17 in the guard interval is the same as the last N _g data 19 in the OFDM useful data, the differential correlation of the received signal is calculated at the receiving end according to the following formula to obtain time and frequency synchronization:

$\mathrm{<span\; class="notranslate">\; P</span>}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; arg</span>}\underset{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; (</span>\underset{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; g</span>}}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; r</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; \&theta;</span>}<span\; class="notranslate">\; ]</span>\mathrm{<span\; class="notranslate">\; r</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; \&theta;</span>}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>$

$\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&epsiv;</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; the\; tan</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; P</span>}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span>$

表示估计的时间同步点，

表示估计的频率偏移值，r[k]为接收信号，N_g in,

represents the estimated time synchronization point,

Indicates the estimated frequency offset value, r[k] is the received signal, N _g

is the length of the guard interval, and N is the number of sampling points of one OFDM symbol.

2) The sending end fills the OFDM symbol with the training sequence, which can be done in the following two ways: a) placing the training sequence 20 in the OFDM guard interval (as shown in Figure 3); b) placing the training sequence 22 in the OFDM guard interval before (as shown in Figure 4). The receiving end performs time synchronization by correlating the received signal with the known training sequence according to the following formula:

$\mathrm{<span\; class="notranslate">\; \&gamma;</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; t</span>}}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}\mathrm{<span\; class="notranslate">\; r</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; \&Center\; Dot;</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 3</span>}<span\; class="notranslate">\; )</span>$

Among them, r[k] is the received signal, N _t is the length of the training sequence, a is the number of sliding points when the received sequence is aligned with the local training sequence, and t[k] is the training sequence.

3) The initiator superimposes the training sequence on the OFDM useful data, see the literature Tufvesson, F.; Edfors, O.; Faulkner, M., "Time and frequency synchronization for OFDM using PN-sequence preambles," Vehicular Technology Conference, 1999.VTC 1999-Fall. IEEE VTS 50th, Volume: 4, 1999, Page(s): 2203-2207.

Place the guard interval as shown in Figure 5, that is, superimpose the training sequence 28 and 30 on the OFDM useful data 29 and 31, and then move 30 and 31 to 26 and 27 to form a guard interval, and the receiving end uses the above two The formula correlates for time synchronization.

However, the above-mentioned OFDM guard interval design methods have shortcomings. The disadvantage of method (1) is that the correlation peak at the receiving end is not obvious, and the frequency offset estimation range is only 1/2 of the OFDM system subcarrier spacing. The disadvantage of method (2) is that the training sequence and OFDM raw data are in the form of time division (or frequency division) multiplexing, resulting in a decrease in data transmission efficiency; in addition, when the position of the training sequence is as shown in Figure 3, it will reduce performance of channel estimation. The disadvantage of method (3) is that the training sequence interferes too much with the data, and the energy efficiency of transmitting data is low.

Contents of the invention

The purpose of the present invention is to provide a method for time and frequency synchronization of an OFDM system, so that the OFDM system has the advantages of good synchronization performance and little interference of training sequences to data.

The innovation of the present invention is: 1) the sending end only places the training sequence within the guard interval of the OFDM symbol, and the training sequence is formed by repeating a selected PN sequence multiple times continuously, and the receiving end performs corresponding detection and processing on this sequence , to realize OFDM synchronization. Since the training sequence is only placed in the guard interval, its interference to the data is smaller than that of the conventional training sequence placement method. At the same time, because the training sequence and OFDM original data are point-to-point weighted superposition, there is no decrease in data transmission efficiency. . 2) The receiving end comprehensively utilizes the correlation between the data in the OFDM guard interval and the data of the last N _g points in the original OFDM data and the correlation introduced by the training sequence, and uses the fusion algorithm to realize time and frequency synchronization, and the performance is better than the traditional method .

The present invention is a method for synchronizing OFDM time and frequency, which is characterized in that the data on N _g points behind the OFDM original data and the training sequence point-to-point weighted superimposition at the originating end constitute a guard interval, and the OFDM original data remains unchanged; the OFDM system The receiving end adopts the fusion algorithm, comprehensively utilizes the correlation between the data in the OFDM guard interval and the data of the last N _g points in the OFDM original data, and the correlation introduced by the training sequence to process the received data to achieve time and frequency synchronization.

According to a kind of OFDM time of the present invention, frequency synchronization method, it is characterized in that it comprises the following steps (as shown in Figure 6):

1. Origin:

1) Select a PN sequence with a length of N _PN , denoted as PN[k], note that the value of PN[k] at this time is in complex form, that is, m[k]∈{1+j, -1-j} ;

2) Place the above-mentioned PN sequence repeatedly repeatedly, cut off the redundant data behind, and form a training sequence with a length of _Ng (as shown in Figure 7), which is recorded as t[k], and its mathematical expression is:

t[k]=m[k mod N _PN ] k ∈ [0, N _g -1] (4)

3) As shown in Figure 6, the training sequence and the data point-to-point weighted superposition of the N _g points behind the OFDM original data are placed in the guard interval; the OFDM original data remains unchanged. Therefore, the mathematical expression of OFDM transmission data is as follows:

$\mathrm{<span\; class="notranslate">\; the\; s</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; =</span>\left[\begin{array}{cc}\sqrt{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; \&rho;</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>\mathrm{<span\; class="notranslate">\; d</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; +</span>\sqrt{\mathrm{<span\; class="notranslate">\; \&rho;</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; ]</span>& \mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; \&Element;</span><span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; g</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ]</span>\\ \mathrm{<span\; class="notranslate">\; d</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; ]</span>& \mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; \&Element;</span><span\; class="notranslate">\; [</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; g</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; g</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ]</span>\end{array}\right.<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 5</span>}<span\; class="notranslate">\; )</span>$

Among them, s[k] represents OFDM transmission data, d[k] represents OFDM useful data, t[k] is the training sequence, ρ represents the weight value, and its physical meaning is that in the guard interval, the energy of the training sequence is relative to the transmission data The normalized value of the energy.

2. Receiver:

1) Coarse time synchronization: The data 33 in the guard interval is the same as the last N _g data 35 in the original OFDM data, and the approximate range of the time synchronization point can be found by using the traditional synchronization method (1) according to the formula (1).

2) Time fine synchronization: within the above range, use the training sequence superimposed in the guard interval to find the precise position of the time synchronization point. The specific method is as follows: let L be the complete number of PN sequences contained in the training sequence, and set a The size of the detection window is N _PN L, and the received data is shifted, and then the correlation is calculated according to the following formula:

$\mathrm{<span\; class="notranslate">\; \&gamma;</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; L</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}<span\; class="notranslate">\; [</span><span\; class="notranslate">\; (</span>\underset{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; PN</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; PN</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; ]</span>\mathrm{<span\; class="notranslate">\; r</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; l</span>}{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; PN</span>}}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; )</span>$

$<span\; class="notranslate">\; \&Center\; Dot;</span>{<span\; class="notranslate">\; (</span>\underset{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; PN</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; PN</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; ]</span>\mathrm{<span\; class="notranslate">\; r</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; PN</span>}}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 6</span>}<span\; class="notranslate">\; )</span>$

Among them, r[k] is the received signal, and a is the number of sliding points of the received sequence relative to the local PN sequence.

3) When γ[k, a] reaches a peak value or exceeds a certain threshold (this threshold can be set according to the actual system), time synchronization is realized.

4) Coarse frequency synchronization: On the basis of the above time synchronization, first use the training sequence superimposed in the guard interval to perform a rough estimation of the frequency offset according to the following formula:

$\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&epsiv;</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; N</span>}}{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;</span>}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; PN</span>}}}<span\; class="notranslate">\; \&Center\; Dot;</span>\mathrm{<span\; class="notranslate">\; arg</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&gamma;</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 7</span>}<span\; class="notranslate">\; )</span>$

5) Fine frequency synchronization: Then use the data 33 in the guard interval to be the same as the last N _g data 35 in the original OFDM data, and use the traditional synchronization method (1) to perform precise frequency offset estimation according to formula (2).

6) According to the above frequency offset estimation, make frequency offset compensation for the received data.

This design approach is based on:

1) Since the training sequence is known, a certain correlation is provided, so it is easy to realize time and frequency synchronization of OFDM;

2) Compared with the method of placing the training sequence in all OFDM symbols, the method of only placing the training sequence in the guard interval adopted by the present invention can reduce the interference of the training sequence to the original data, and improve energy efficiency simultaneously;

3) Since the training sequence and OFDM original data are point-to-point weighted superposition in the guard interval, there is no decrease in data transmission efficiency;

4) The weighted value ρ during point-to-point weighted superposition can be different for different OFDM symbols, but within the same OFDM symbol, the value of ρ remains unchanged. In order to ensure accurate synchronization, the signal power of the training sequence components in the guard interval can be increased by increasing the value of ρ in the acquisition phase, and the value of ρ can be decreased in the tracking phase. Adjusting ρ allows for optimal time and frequency synchronization. When ρ=1, all guard interval part signals are replaced by PN sequence signals; when ρ=0, the guard interval structure is the same as the conventional guard interval (method one mentioned above, that is, the guard interval signal is part of the OFDM useful signal) The structure is the same. The selection of ρ value and the design of PN sequence are one of the key technologies in the design of this guard interval structure.

5) When the receiving end uses the training sequence for time fine synchronization and frequency coarse synchronization, it first correlates the received data segment with the local PN sequence, and then performs differential correlation calculation. This processing method can offset the influence of the channel impulse response , thereby improving synchronization performance.

6) Since the present invention comprehensively utilizes the correlation between the data 33 in the OFDM guard interval and the data 35 of the rear _Ng point in the OFDM raw data and the correlation introduced by the training sequence, both time and frequency synchronization are performed (this is called fusion algorithm), thus outperforming traditional methods.

7) Since the training sequence signal is known at the receiving end, it can also be used for channel estimation or other purposes.

The present invention has the following characteristics:

1. The data in the OFDM guard interval is formed by superimposing the data on N _g points behind the OFDM useful signal and the point-to-point weighted training sequence, and the original OFDM data remains unchanged;

2. The weight value of point-to-point weighted superposition is ρ, and the value of ρ can be different for different OFDM symbols, but within the same OFDM symbol, the value of ρ remains unchanged. By adjusting this weight, a trade-off is made between the effectiveness of the transmission and the reliability of the training sequence for synchronization estimation;

3. The training sequence used for synchronization is composed of multiple identical PN sequences with good correlation characteristics placed continuously or sequences with other characteristics (as shown in Figure 7);

4. When the receiving end performs time and frequency synchronization, a fusion algorithm is used, that is, the correlation between the data 33 in the OFDM guard interval and the data 35 at the last N _g points in the original OFDM data and the correlation introduced by the training sequence are comprehensively utilized.

5. When the receiving end uses the training sequence to perform time fine synchronization and frequency coarse synchronization calculation, it first correlates the received data segment with the local PN sequence, and then performs differential correlation calculation.

Description of drawings

Figure 1 is a general OFDM system block diagram

In the figure, 1 is the serial-to-parallel conversion module, 2 is the mapping module, 3 is the IFFT module, 4 is the parallel-serial conversion module, 5 is the protection time slot addition module, 6 is the D/A low-pass filter module, and 7 is the up-conversion module , 8 is the channel module, 9 is the down-conversion module, 10 is the A/D low-pass filter module, 11 is the synchronization module, 12 is the deprotection time slot module, 13 is the FFT module, 14 is the first-order equalization module, and 15 is the mapping module, 16 is a parallel module;

Fig. 2 is the structure of the OFDM symbol adopting conventional synchronization method (1)

In the figure, the data 19 of N _g points in the back of the OFDM original data is copied to the front 17 of OFDM useful data to form a guard interval with a length of N _g , and the original OFDM data 18 and 19 remain unchanged. The correlation performance of the receiver, that is to say, the sharpness of the objective function calculated by the receiver for time synchronization search cannot be guaranteed, so the time synchronization performance of this method is not good;

Fig. 3 is the structure of the OFDM symbol that places the training sequence in the guard interval of OFDM using the conventional synchronization method (2)

In the figure, the data 20 in the guard interval of the OFDM transmission symbol is a training sequence for synchronization, does not contain any OFDM original data, and the OFDM original data 21 remains unchanged;

Fig. 4 adopts the structure of the OFDM symbol that the training sequence 22 is placed before the guard interval of OFDM for adopting the conventional synchronization method (2)

In the figure, the data 25 of the last N _g points of the OFDM raw data is first copied to the front 23 of the OFDM raw data, and then the training sequence 22 for synchronization is placed in front of 23, and other OFDM raw data 24, 25 are not Obviously, because the training sequence occupies a part of the transmission time, the data transmission rate drops;

Fig. 5 is the structure of the OFDM symbol adopting conventional synchronization method (3)

In the figure, the training sequence 30 used for synchronization is first superimposed on the data 31 of the last N _g points of the OFDM raw data, and then the data 30 and 31 of the last N _g points in the obtained sequence are copied to the front of the OFDM 26 , 27, constituting the guard interval, because each OFDM raw data participates in the point-to-point weighted superposition with the training sequence data, so it is greatly interfered by the training sequence;

Figure 6 is the structure of the OFDM symbol using the synchronization method described in this patent

In the figure, the sequence 33 is a copy of the data 35 on N _g points behind the OFDM original data, and the point-to-point weighted superposition of the sequence 33 and the training sequence 32 constitutes a guard interval, and the OFDM original data 34 and 35 remain unchanged, wherein the training sequence 32 The content of is shown in Figure 7, since the training sequence data is only superimposed with the OFDM original data in the guard interval, there is no interference to other parts of the OFDM original data;

Figure 7 is the internal structure of the training sequence in the guard interval of the OFDM symbol described in this patent

In the figure, the training sequence is composed of a certain PN sequence placed continuously, among which, 36, 37, 38, ..., 39 are copies of the same PN sequence, and 40 may be a complete PN sequence, or it may be truncated The PN sequence, the length of the training sequence finally obtained is N _g , and the good correlation characteristics of the PN sequence can ensure the time-frequency synchronization performance of the method described in this patent;

Fig. 8 is a block diagram of the originating implementation steps described in this patent

Fig. 9 is a block diagram of the receiving end implementation steps described in this patent

It should be noted that, in the accompanying drawings of the specification, N is the number of FFT points, N _g is the number of guard interval points, and T _s is the sampling interval.

Detailed ways

The implementation steps of this patent under a specific OFDM configuration are given below. It should be noted that the parameters in the following examples do not affect the generality of this patent.

It is assumed that the OFDM useful symbol length is N=4096, and the guard interval length is N _g =512. Taking ρ=0.5 means that the energy of the training sequence in the guard interval accounts for 50% of the total energy. The PN sequence is selected as an m-sequence whose period is N _PN =31, denoted as PN[i]i∈[0,30], then the number of repetitions of the m-sequence in the training sequence is L=N _g /N _PN =512/31= 17.

The transmitter repeats the m-sequence 17 times in a row according to Formula 4 (the last m-sequence will be truncated) to form a training sequence, and performs point-to-point weighted superposition with the original OFDM data according to Formula 5 before transmitting it.

The receiving end first performs rough time synchronization according to Formula 1, determines the approximate position of the time synchronization point, and then performs fine time synchronization according to Formula 6, calculates the precise position of the time synchronization point, and obtains time synchronization, namely:

$\mathrm{<span\; class="notranslate">\; \&gamma;</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; 16</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}<span\; class="notranslate">\; [</span><span\; class="notranslate">\; (</span>\underset{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; 31</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; PN</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; ]</span>\mathrm{<span\; class="notranslate">\; r</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; l</span>}{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; PN</span>}}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; )</span>$

$<span\; class="notranslate">\; \&Center\; Dot;</span>{<span\; class="notranslate">\; (</span>\underset{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; 31</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; PN</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; ]</span>\mathrm{<span\; class="notranslate">\; r</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; PN</span>}}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; ]</span>$

Assuming that when a is 18, γ[k, a], that is, γ[k, 18] reaches a peak value, and time synchronization is achieved.

After time synchronization is obtained, the frequency offset is roughly estimated according to formula 7, namely:

$\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&epsiv;</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 4096</span>}}{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;</span>}<span\; class="notranslate">\; \&Center\; Dot;</span>\mathrm{<span\; class="notranslate">\; 31</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>\mathrm{<span\; class="notranslate">\; arg</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; r</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 18</span>}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; )</span>$

Then, start frequency offset fine estimation according to formula 2. Finally, frequency offset compensation is performed to obtain frequency synchronization.

Claims (3)

1, a kind of OFDM time, frequency synchronous method, the sender processing step of this synchronous method is: Step 1 Select a PN sequence with a length of N _PN , denoted as PN[k], and the value of PN[k] at this time is in complex form, that is, m[k]∈{1+j, -1-j}; Step 2 Repeat the above PN sequence several times continuously, cut off the redundant data behind, and form a training sequence with a length of _Ng , which is recorded as t[k], and its mathematical expression is: t[k]=m[k mod N _PN ] k∈[0,N _g -1] Step 3 After the weighted point-to-point superimposition of the training sequence and the data on the N _g points behind the OFDM original data, place them in the guard interval; the original OFDM data remains unchanged, so the mathematical expression of the OFDM transmitted data is as follows: $\mathrm{<span\; class="notranslate">\; the\; s</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; =</span>\left\{\begin{array}{cc}\sqrt{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; \&rho;</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>\mathrm{<span\; class="notranslate">\; d</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; +</span>\sqrt{\mathrm{<span\; class="notranslate">\; \&rho;</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; ]</span>& \mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; \&Element;</span><span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; g</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ]</span>\\ \mathrm{<span\; class="notranslate">\; d</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; ]</span>& \mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; \&Element;</span><span\; class="notranslate">\; [</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; g</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; g</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ]</span>\end{array}\right.$ The steps of the receiving end processing of the OFDM system are: Step 4 Coarse time synchronization: use the data (33) in the guard interval to be the same as the last N _g data (35) in the OFDM original data, and calculate the differential correlation of the received signal at the receiving end, and the coarse time synchronization can be achieved according to the following formula Synchronize; $\mathrm{<span\; class="notranslate">\; P</span>}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; arg</span>}\underset{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; (</span>\underset{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; g</span>}}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; r</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; \&theta;</span>}<span\; class="notranslate">\; ]</span>\mathrm{<span\; class="notranslate">\; r</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; \&theta;</span>}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; )</span>$ Step 5 Time fine synchronization: Within the above range, use the training sequence superimposed in the guard interval to find the precise position of the time synchronization point. The specific method is as follows: Let L be the complete number of PN sequences contained in the training sequence, and set a The size of the detection window is N _PN L, and the received data is shifted, and then the correlation is calculated according to the following formula: $\mathrm{<span\; class="notranslate">\; \&gamma;</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; L</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}<span\; class="notranslate">\; [</span><span\; class="notranslate">\; (</span>\underset{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; PN</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; PN</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; ]</span>\mathrm{<span\; class="notranslate">\; r</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; l</span>}{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; PN</span>}}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; )</span>$ $<span\; class="notranslate">\; \&Center\; Dot;</span>{<span\; class="notranslate">\; (</span>\underset{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; PN</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; PN</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; ]</span>\mathrm{<span\; class="notranslate">\; r</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; PN</span>}}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; ]</span>$ Step 6 When r[k, a] reaches the peak value or exceeds a certain threshold value set according to the actual system, time synchronization is realized; Step 7 Coarse frequency synchronization: On the basis of the above time synchronization, first use the training sequence superimposed in the guard interval to perform a rough estimation of the frequency offset according to the following formula: $\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&epsiv;</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; N</span>}}{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;</span>}<span\; class="notranslate">\; \&CenterDot;</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; PN</span>}}}<span\; class="notranslate">\; \&CenterDot;</span>\mathrm{<span\; class="notranslate">\; arg</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&gamma;</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; )</span>$ Step 8 Fine frequency synchronization: Then use the data in the guard interval to be the same as the last N _g data in the original OFDM data, and use the synchronization method in step 4 to perform precise frequency offset estimation according to the following formula: $\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&epsiv;</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;</span>}}<span\; class="notranslate">\; \&CenterDot;</span>{\mathrm{<span\; class="notranslate">\; the\; tan</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; P</span>}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span>$ Step 9 Make frequency offset compensation for the received data according to the above frequency offset estimation. 2. According to a kind of OFDM time and frequency synchronization method according to claim 1, it is characterized in that the weighted value ρ when described point-to-point weighted superposition can be different to different OFDM symbols, but in the same OFDM symbol, ρ Constant value of : By adjusting this weight, a trade-off is made between the effectiveness of the transmission and the reliability of the training sequence for synchronization estimation. 3, according to a kind of OFDM time according to claim 1, frequency synchronization method, it is characterized in that in the step that described OFDM system receiving terminal processes, receiving terminal is when utilizing training sequence to carry out time fine synchronization and frequency coarse synchronization calculation, Firstly, the received data segment is correlated with the local PN sequence, and then the differential correlation calculation is performed.

Images