METHOD, RECEIVER AND SYSTEM FOR SIGNAL SYNCHRONIZATION

• US 20140321450A1
• Filed: 03/20/2014
• Published: 10/30/2014
• Est. Priority Date: 03/21/2013
• Status: Active Grant

• Current Assignee: Beijing Lenovo Software Ltd (Lenovo Group Ltd.), Lenovo (Beijing) Limited (Lenovo Group Ltd.)
• Original Assignee: Beijing Lenovo Software Ltd (Lenovo Group Ltd.), Lenovo (Beijing) Limited (Lenovo Group Ltd.)
• Inventors: Zhang, Li, Wang, Jie
• Granted Patent: US 9,337,994 B2
• Time in Patent Office: Days
• US Class Current: 370/350
• CPC Class Codes: H04J 11/0073 Acquisition of primary synchronisation channel, e.g. detection of cell-ID within cell-ID group
• CPC Class Codes: H04J 11/0076 Acquisition of secondary synchronisation channel, e.g. detection of cell-ID group
• CPC Class Codes: H04L 27/26 Systems using multi-frequency codes H04L27/32 takes precedence
• CPC Class Codes: H04L 27/2659 Coarse or integer frequency offset determination and synchronisation
• CPC Class Codes: H04L 27/266 Fine or fractional frequency offset determination and synchronisation
• CPC Class Codes: H04L 27/2665 Fine synchronisation, e.g. by positioning the FFT window
• CPC Class Codes: H04L 5/0007 the frequencies being orthogonal, e.g. OFDM(A), DMT
• CPC Class Codes: H04L 7/0016 correction of synchronization errors
• CPC Class Codes: H04W 56/0035 detecting errors in frequency or phase
• CPC Class Codes: H04W 56/0085 detecting a given structure in the signal

Abstract

In a method for time domain joint synchronization in downlink, a sequence to be analyzed is obtained from a received time domain signal; the sequence to be analyzed is down-sampled by a factor of α to obtain a down-sampled sequence; an initial timing estimation is performed based on the down-sampled sequence; a frequency offset compensation is applied to the sequence to be analyzed based on an Integral Frequency Offset (IFO) estimation value {circumflex over (ε)} obtained by the initial timing estimation; a Fractional Frequency Offset (FFO) estimation value {circumflex over (ε)}_f is obtained; a frequency offset compensation is applied to the sequence to be analyzed based on the IFO estimation value {circumflex over (ε)} and the FFO estimation value {circumflex over (ε)}_f; and a further timing estimation is performed based on a position interval estimated using an initial timing position and a Primary Synchronization Sequence (PSS) to obtain a fine timing position.

Claims

1. A method for signal synchronization, the method being applied in a receiver and comprising:

obtaining a sequence to be analyzed y from a received time domain signal;

down-sampling the sequence to be analyzed y by a factor of α

to obtain a down-sampled sequence and performing an initial timing estimation based on the down-sampled sequence and n different predetermined Integral Frequency Offset (IFO) candidate values to determine an initial timing position T_coarse, an IFO estimation value {circumflex over (ε

)} and a Primary Synchronization Sequence (PSS) corresponding to the sequence to be analyzed y;

applying an initial frequency offset compensation to the sequence to be analyzed y based on the IFO estimation value {circumflex over (ε

)} and performing an initial frequency offset estimation based on the initial timing position T_coarse and the determined PSS to obtain a Fractional Frequency Offset (FFO) estimation value {circumflex over (ε

)}_f; and

applying a frequency offset compensation to the sequence to be analyzed y based on the IFO estimation value {circumflex over (ε

)} and the FFO estimation value {circumflex over (ε

)}_f and performing a further timing estimation based on the initial timing position T_coarse and the determined PSS to obtain a fine timing position T_fine,

wherein n is a positive integer equal to or larger than 1.

2. The method of claim 1, further comprising, after the fine timing position T_fine is obtained:

extracting a secondary sequence segment to be analyzed from the sequence to be analyzed y based on the fine timing position T_fine, and

applying a frequency offset compensation to the secondary sequence segment to be analyzed based on the IFO estimation value {circumflex over (ε

)} and the FFO estimation value {circumflex over (ε

)}_f and determining a Secondary Synchronization Sequence (SSS) corresponding to the secondary sequence segment to be analyzed and a timing position of a wireless data frame using a cross-correlation calculation,

wherein m is a positive integer equal to or larger than 1.

3. The method of claim 2, further comprising, after applying a frequency offset compensation to the secondary sequence segment to be analyzed based on the IFO estimation value {circumflex over (ε

)} and the FFO estimation value {circumflex over (ε

)}_f and determining a Secondary Synchronization Sequence (SSS) corresponding to the secondary sequence segment to be analyzed and a timing position of a wireless data frame using a cross-correlation calculation;

performing a further frequency offset estimation based on the secondary sequence segment to be analyzed and the SSS corresponding to the secondary sequence segment to be analyzed to regenerate a FFO estimation value {circumflex over (ε

)}_f and using the IFO estimation value and the regenerated FFO estimation value {circumflex over (ε

)}_f as final frequency offset estimation values.

4. The method of claim 1, wherein said obtaining a sequence to be analyzed y from a received time domain signal comprises:

extracting from the received time domain signal a sequence having a sampling rate equal to a maximum bandwidth sampling rate and a length of N_buffer using a sliding window; and

applying a narrow-band filtering to the sequence having the length of N_buffer to obtain the sequence to be analyzed y that is located at a center position of a transmission bandwidth and has a bandwidth of a predetermined value,

wherein N_buffer belongs to (half of wireless frame length, half of wireless frame length+one symbol length) and the predetermined value belongs to [1.08 MHz, 1.25 MHz].

5. The method of claim 1, wherein said down-sampling the sequence to be analyzed y by a factor of α

to obtain a down-sampled sequence and performing an initial timing estimation based on the down-sampled sequence and n different predetermined Integral Frequency Offset (IFO) candidate values to determine an initial timing position T_coarse, an IFO estimation value {circumflex over (ε

)} and a Primary Synchronization Sequence (PSS) corresponding to the sequence to be analyzed y comprises;

down-sampling the sequence to be analyzed y by the factor of α

to obtain the down-sampled sequence, where α

is 2ⁿ;

applying a frequency offset compensation to the down-sampled sequence based on the n different predetermined IFO candidate values to obtain n IFO compensated down-sampled sequences;

performing a cross-correlation calculation based on the n IFO compensated down-sampled sequences and k time domain PSSs generated at the receiver to obtain k*n cross-correlation result sequences;

determining a maximum element max, in each of the cross-correlation result sequences and a position d, corresponding to the maximum element max_i;

selecting one of the k*n maximum elements max, that has a largest value, denoted as a largest element max;

detecting whether the largest element max is larger than a predetermined threshold P_th; and

determining, when it is detected that the largest element max is larger than the predetermined P_th, the position d, corresponding to the largest element max as the initial timing position T_coarse, the predetermined IFO candidate value corresponding to the cross-correlation result sequence to which the largest element max belongs as the IFO estimation value {circumflex over (ε

)} and the PSS corresponding to the cross-correlation result sequence to which the largest element max belongs as the PSS corresponding to the sequence to be analyzed y,

wherein k is a positive integer equal to or larger than 1.

6. The method of claim 5, wherein said performing a cross-correlation calculation based on the n IFO compensated down-sampled sequences and k time domain PSSs generated at the receiver to obtain k*n cross-correlation result sequences comprises:

assuming the IFO compensated down-sampled sequence as y_compensate and the time domain PSS as pss_time_downsampling, and performing the cross-correlation calculation in accordance with the following block-wise cross-correlation equation;

7. The method of claim 1, wherein said applying a frequency offset compensation to the sequence to be analyzed y based on the IFO estimation value {circumflex over (ε

)} and performing an initial frequency offset estimation based on the initial timing position T_coarse and the determined PSS to obtain a Fractional Frequency Offset (FFO) estimation value {circumflex over (ε

)}_f comprises;

applying a frequency offset compensation to the sequence to be analyzed y based on the IFO estimation value {circumflex over (ε

)} to obtain a frequency offset compensated sequence to be analyzed y;

extracting from the frequency offset compensated sequence to be analyzed y a primary sequence segment to be analyzed Rx_rss based on the initial timing position T_coarse;

performing an initial frequency offset estimation based on the primary sequence segment to be analyzed Rx_rss and the determined PSS LocalPss to obtain the FFO estimation value {circumflex over (ε

)}_f in accordance with;

8. The method of claim 1, wherein said applying a frequency offset compensation to the sequence to be analyzed y based on the IFO estimation value {circumflex over (ε

)} and the FFO estimation value {circumflex over (ε

)}_f and performing a further timing estimation based on the initial timing position T_coarse and the determined PSS to obtain a fine timing position T_fine comprises;

applying a frequency offset compensation to the sequence to be analyzed y based on the IFO estimation value {circumflex over (ε

)} and the FFO estimation value {circumflex over (ε

)}_f to obtain a frequency offset compensated sequence to be analyzed y;

determining a position interval [T_coarse−

R, T_coarse+R] from the frequency offset compensated sequence to be analyzed y based on the initial timing position T_coarse;

dividing the frequency offset compensated sequence to be analyzed y into 2R+1 data sequences Rx_Search based on the position interval;

performing a cross-correlation calculation based on the 2R+1 data sequences Rx_Search and the determined PSS LocalPss to obtain 2R+1 cross-correlation result sequences;

determining a maximum element pmax_i in each of the cross-correlation result sequences and a position pd_i corresponding to the maximum element pmax_i;

selecting one of the 2R+1 maximum elements pmax_i that has a largest value, denoted as a largest element pmax; and

determining the position pd, corresponding to the maximum element pmax as the fine timing position T_fine,

wherein R is a positive integer equal to or larger than 1.

9. The method of claim 2, wherein said extracting a secondary sequence segment to be analyzed from the sequence to be analyzed y based on the fine timing position T_fine comprises:

calculating a start position T_SSS of the secondary sequence segment to be analyzed based on a timing relationship between the PSS and the SSS and the fine timing position T_fine in accordance with;

T_SSS=T_fine−

L_symbol−

L_CP;

determining whether the start position T_SSS is located within the sequence to be analyzed y;

extracting the secondary sequence segment to be analyzed based on the start position T_SSS when it is determined that the start position T_SSS is located within the sequence to be analyzed y;

recalculating, when it is determined that the start position T_SSS is located outside the sequence to be analyzed y, the start position T_SSS in accordance with;

T_SSS=T_fine−

L_symbol−

L_CP+L<sub>half—

of—

wireless—

frame</sub>;

extracting the secondary sequence segment to be analyzed based on the recalculated start position T_SSS,

where T_fine is a sequence number of the start position of the PSS corresponding to the sequence to be analyzed y, L_symbol is a length of a symbol, L_CP is a length of a Cyclic Prefix (CP), and L<sub>half—

of—

wireless—

frame</sub> is a length of half of a wireless data frame.

10. The method of claim 9, wherein said applying a frequency offset compensation to the secondary sequence segment to be analyzed based on the IFO estimation value and the FFO estimation value {circumflex over (ε

)}_f and determining a Secondary Synchronization Sequence (SSS) corresponding to the secondary sequence segment to be analyzed and a timing position of a wireless data frame using a cross-correlation calculation comprises;

applying a frequency offset compensation to the secondary sequence segment to be analyzed based on the IFO estimation value {circumflex over (ε

)} and the FFO estimation value {circumflex over (ε

)}_f to obtain a frequency offset compensated secondary sequence segment to be analyzed Rx_Sss;

performing a cross-correlation calculation based on the frequency offset compensated secondary sequence segment to be analyzed Rx_Sss and m time domain SSSs generated at the receiver to obtain m cross-correlation result sequences;

determining a maximum element smax_i in each of the cross-correlation result sequences and a position sd_i corresponding to the maximum element smax_i;

selecting one of the m maximum elements smax_i that has a largest value, denoted as a largest element smax;

determining the SSS corresponding to the largest element smax as the SSS corresponding to the secondary sequence segment to be analyzed; and

determining the timing position of the wireless data frame based on a type of the SSS corresponding to the secondary sequence segment to be analyzed and the position sd_i corresponding to the largest element smax,

wherein m is a positive integer equal to or larger than 1.

11. The method of claim 3, wherein said performing a further frequency offset estimation based on the secondary sequence segment to be analyzed and the SSS corresponding to the secondary sequence segment to be analyzed to regenerate a FFO estimation value {circumflex over (ε

)}_f and using the IFO estimation value {circumflex over (ε

)} and the regenerated FFO estimation value {circumflex over (ε

)}_f as final frequency offset estimation values comprises;

assuming the secondary sequence segment to be analyzed as Rx_Sss and the SSS corresponding to the secondary sequence segment to be analyzed as LocalSss and recalculating the FFO estimation value {circumflex over (ε

)}_f in accordance with;

12. A receiver comprising:

a sequence obtaining module configured to obtain a sequence to be analyzed y from a received time domain signal;

an initial timing estimation module configured to down-sample the sequence to be analyzed y as obtained by the sequence obtaining module by a factor of α

to obtain a down-sampled sequence and perform an initial timing estimation based on the down-sampled sequence and n different predetermined Integral Frequency Offset (IFO) candidate values to determine an initial timing position T_coarse, an IFO estimation value {circumflex over (ε

)} and a Primary Synchronization Sequence (PSS) corresponding to the sequence to be analyzed y;

an initial frequency offset estimation module configured to apply an initial frequency offset compensation to the sequence to be analyzed y as obtained by the sequence obtaining module based on the IFO estimation value {circumflex over (ε

)} as determined by the initial timing estimation module and perform an initial frequency offset estimation based on the initial timing position T_coarse as determined by the initial timing estimation module and the determined PSS to obtain a Fractional Frequency Offset (FFO) estimation value {circumflex over (ε

)}_f; and

a further timing estimation module configured to apply a frequency offset compensation to the sequence to be analyzed y as obtained by the sequence obtaining module based on the IFO estimation value {circumflex over (ε

)} as determined by the initial timing estimation module and the FFO estimation value {circumflex over (ε

)}_f as determined by the initial frequency offset estimation module and perform a further timing estimation based on the initial timing position T_coarse as determined by the initial timing estimation module and the determined PSS to obtain a fine timing position T_fine,

wherein n is a positive integer equal to or larger than 1.

13. The receiver of claim 12, further comprising:

a secondary sequence segment extraction module configured to extract a secondary sequence segment to be analyzed from the sequence to be analyzed y based on the fine timing position T_fine as determined by the further timing estimation module; and

a Secondary Synchronization Sequence (SSS) detection module configured to apply a frequency offset compensation to the secondary sequence segment to be analyzed as extracted by secondary sequence segment extraction module the based on the IFO estimation value {circumflex over (ε

)} as determined by the initial timing estimation module and the FFO estimation value {circumflex over (ε

)}_f as determined by the initial frequency offset estimation module and determine a Secondary Synchronization Sequence (SSS) corresponding to the secondary sequence segment to be analyzed and a timing position of a wireless data frame using a cross-correlation calculation,

wherein m is a positive integer equal to or larger than 1.

14. The receiver of claim 13, further comprising:

a further frequency offset estimation module configured to perform a further frequency offset estimation based on the secondary sequence segment to be analyzed as extracted by secondary sequence segment extraction module and the SSS corresponding to the secondary sequence segment to be analyzed as determined by the SSS detection module to regenerate a FFO estimation value {circumflex over (ε

)}_f and use the IFO estimation value {circumflex over (ε

)} as determined by the initial timing estimation module and the regenerated FFO estimation value {circumflex over (ε

)}_f as final frequency offset estimation values.

15. The receiver of claim 12, wherein the sequence obtaining module comprises:

a sliding extraction unit configured to extract from the received time domain signal a sequence having a sampling rate equal to a maximum bandwidth sampling rate and a length of N_buffer using a sliding window; and

a narrow-band filtering unit configured to apply a narrow-band filtering to the sequence having the length of N_buffer as extracted by the sliding extraction unit to obtain the sequence to be analyzed y that is located at a center position of a transmission bandwidth and has a bandwidth of a predetermined value,

wherein N_buffer belongs to (half of wireless frame length, half of wireless frame length+one symbol length) and the predetermined value belongs to [1.08 MHz, 1.25 MHz].

16. The receiver of claim 12, wherein the initial timing estimation module comprises:

a down-sampling unit configured to down-sample the sequence to be analyzed y by the factor of α

to obtain the down-sampled sequence, where a is 2ⁿ;

a candidate frequency offset compensation unit configured to apply a frequency offset compensation to the down-sampled sequence as obtained by the down-sampling unit based on the n different predetermined IFO candidate values to obtain n IFO compensated down-sampled sequences;

a cross-correlation calculation unit configured to perform a cross-correlation calculation based on the n IFO compensated down-sampled sequences as obtained by the candidate frequency offset compensation unit and k time domain PSSs generated at the receiver to obtain k*n cross-correlation result sequences;

a position determination unit configured to determine a maximum element max_i in each of the cross-correlation result sequences as obtained by the cross-correlation unit and a position d, corresponding to the maximum element max_i;

a maximum selection unit configured to select one of the k*n maximum elements max_i as determined by the position determination unit that has a largest value, denoted as a largest element max;

a largest detection unit configured to detect whether the largest element max is larger than a predetermined threshold P_th; and

an initial estimation unit configured to determine, when the largest detection unit detects that the largest element max is larger than the predetermined P_th, the position d_i corresponding to the largest element max as the initial timing position T_coarse) the predetermined IFO candidate value corresponding to the cross-correlation result sequence to which the largest element max belongs as the IFO estimation value and the PSS corresponding to the cross-correlation result sequence to which the largest element max belongs as the PSS corresponding to the sequence to be analyzed y,

wherein k is a positive integer equal to or larger than 1.

17. The receiver of claim 16, wherein the cross-correlation calculation unit is configured to assume the IFO compensated down-sampled sequence as y_compensate and the time domain PSS as pss_time_downsampling, and perform the cross-correlation calculation in accordance with the following block-wise cross-correlation equation:

18. The receiver of claim 12, wherein the initial frequency offset estimation module comprises:

an integral compensation unit configured to apply a frequency offset compensation to the sequence to be analyzed y based on the IFO estimation value {circumflex over (ε

)} to obtain a frequency offset compensated sequence to be analyzed y;

a primary sequence segment extraction unit configured to extract from the frequency offset compensated sequence to be analyzed y as obtained by the integral compensation unit a primary sequence segment to be analyzed Rx_rss based on the initial timing position T_coarse;

an initial frequency offset estimation unit configured to perform an initial frequency offset estimation based on the primary sequence segment to be analyzed Rx_rss as extracted by the primary sequence segment extraction unit and the determined PSS LocalPss to obtain the FFO estimation value {circumflex over (ε

)}_f in accordance with;

19. The receiver of claim 12, wherein the further timing estimation module comprises:

a fractional compensation unit configured to apply a frequency offset compensation to the sequence to be analyzed y based on the IFO estimation value and the FFO estimation value to to obtain a frequency offset compensated sequence to be analyzed y;

an interval determination unit configured to determine a position interval [T_coarse−

R, T_coarse+R] from the frequency offset compensated sequence to be analyzed y as obtained by the fractional compensation unit based on the initial timing position T_coarse;

a sequence division unit configured to divide the frequency offset compensated sequence to be analyzed y into 2R+1 data sequences Rx_Search based on the position interval as determined by the interval determination unit;

a cross-correlation calculation unit configured to perform a cross-correlation calculation based on the 2R+1 data sequences Rx_Search and the determined PSS LocalPss to obtain 2R+1 cross-correlation result sequences;

an element determination unit configured to determine a maximum element pmax_i in each of the cross-correlation result sequences as obtained by the cross-correlation calculation unit and a position pd_i corresponding to the maximum element pmax_i;

an element selection unit configured to select one of the 2R+1 maximum elements pmax_i that has a largest value, denoted as a largest element pmax; and

a further timing estimation unit configured to determine the position pd_i corresponding to the maximum element pmax as the fine timing position T_fine;

wherein R is a positive integer equal to or larger than 1.

20. The receiver of claim 13, wherein the secondary sequence segment extraction module comprises:

a first calculation unit configured to calculate a start position T_SSS of the secondary sequence segment to be analyzed based on a timing relationship between the PSS and the SSS and the fine timing position T_fine in accordance with;

T_SSS=T_fine−

L_symbol−

L_CP;

a position determination unit configured to determine whether the start position T_SSS is located within the sequence to be analyzed y;

a first extraction unit configured to extract the secondary sequence segment to be analyzed based on the start position T_SSS when it is determined that the start position T_SSS is located within the sequence to be analyzed y;

a second calculation unit configured to recalculate, when it is determined that the start position T_SSS is located outside the sequence to be analyzed y, the start position T_SSS in accordance with;

T_SSS=T_fine−

L_symbol−

L_CP+L<sub>half—

of—

wireless—

frame</sub>;

a second extraction unit configured to extract the secondary sequence segment to be analyzed based on the recalculated start position T_SSS,

where T_fine is a sequence number of the start position of the PSS corresponding to the sequence to be analyzed y, L_symbol is a length of a symbol, L_CP is a length of a Cyclic Prefix (CP), and L<sub>half—

of—

wireless—

frame</sub> is a length of half of a wireless data frame.