TWI507064B - Neighbor cell search method - Google Patents

Description

Neighbor cell search method

The present invention relates to a method for decoding a cell identifier, and more particularly to a method for decoding a neighbor cell identification code.

In the traditional peak-chastion mobile communication system, in order to achieve time and frequency synchronization between the user equipment (User Equipment, UE) and the base station (eNB), when the user equipment starts, preliminary Cell search. Moreover, the user equipment needs to perform a cell search procedure to obtain the cell identification code of the base station before accessing the base station. Taking the 3rd Generation Partnership Project Long Term Evolution (3GPP LTE) communication system as an example, in each cell, two kinds of synchronization signals are broadcasted to allow the user equipment to perform cell search. program. One is a Primary Synchronization Signal (PSS) and the other is a Secondary Synchronization Signal (SSS). In the downlink system, the user equipment uses time and frequency in addition to PSS and SSS. In the above synchronization, it is also necessary to use these two kinds of synchronization signals to obtain CELL Identities (CELL ID) information.

There are 504 cell identifiers in the LTE system, and the cell identifier is defined by a group ID and a sector ID. The mathematical expression of the cell identifier is . among them, For the group ID, {0,1,...,167}, For the section identification code, {0,1,2}. The above PSS is a Zadoff-Chu sequence, which carries the segment identification code. Information. The above SSS is constructed using two Maximum Length (ML) sequences, wherein the two maximum length sequences are generated by two cyclic shifts ( m ₀ , m ₁ ). Group ID The range is 0 to 167, and corresponds to a unique set of cyclic shifts ( m ₀ , m ₁ ), so that the SSS carries the group identifier Information.

In addition, in order to maintain the connection network, in order to maintain the connection network, in addition to searching for the serving cell currently located, it is also necessary to periodically search for the neighbor cell to enable the mobile device. The user device performs a fast and efficient handoff. However, the signal strength from the neighboring cell base station is generally low, and the signal sent by the current home cell base station can cause serious interference, allowing the user equipment to search for neighboring cells. It is quite difficult.

In the traditional coherent detection method, the channel is estimated by using the received primary synchronization signal PSS, and then the channel effect in the received signal and the signal component in the current serving cell are eliminated, and finally, Segment and group ID of neighboring cells Detection. However, since the received primary synchronization signal PSS contains a lot of interference, the estimated channel response is not accurate enough and the jitter is very severe, thereby causing the detection accuracy to decrease. Therefore, before the detection of the neighboring cells, the receiving end usually uses a channel smoothing filter to perform a signal processing on the estimated channel response, and further increases the receiving end. Computation.

An object of the present invention is to provide a non-coherent neighbor cell search method, which parses out the cell identification code of the current serving cell and the identification code of the neighboring base station from the received synchronization signal, thereby helping the user device to reach the mobile device when moving. Quick hand change function.

In view of the above, the present invention provides a non-coherent neighbor cell search method, which is applicable to a user equipment, wherein the user equipment communicates through a current cell, which is non-coherent. The neighbor cell search method includes: decoding an identifier of a current cell; extracting a component on a kth subcarrier and a k +1th subcarrier in a received signal; multiplying a component of the k +1th subcarrier in the received signal Conjugating a component of the k +1th subcarrier in the previous local synchronization signal to obtain a first product, wherein the local synchronization signal is a synchronization signal corresponding to the cell identifier of the current cell; The component of the kth subcarrier is multiplied by the conjugate of the component of the kth subcarrier in the local synchronization signal to obtain a second product; the difference between the first product and the second product is calculated to obtain a composite signal; The composite signal is correlated with all possible identification codes of neighboring cells to obtain an identification code of the neighboring cell.

According to the non-coherent neighbor cell search method according to the preferred embodiment of the present invention, the step of performing a correlation operation on all possible identification codes of the neighboring cells by the synthesized signal to obtain the identification code of the neighboring cell includes: The operation of synthesizing the signal provides a reference sequence function, denoted as Y ; substitutes M possible identification codes into the reference sequence function Y , wherein the xth identification code is substituted into the reference sequence function Y to obtain Y ( x ); A correlation value operation obtains M correlation values, wherein the Y ( x ) corresponding to the xth possible identification code is correlated with the synthesized signal to obtain the xth correlation value; and the M correlation values are found out a maximum correlation value; and estimating the identification code corresponding to the maximum correlation value as an identification code of the neighboring cell. Where x and M are positive integers.

The spirit of the present invention is to eliminate the signal component of the current serving cell base station in the synchronization signal by generating a synthetic signal mathematical operation, leaving the signal component of the adjacent cell base station, thereby reducing the identification code of the adjacent cell. The error rate at the time of detection.

The above and other objects, features and advantages of the present invention will become more <RTIgt;

S910~S980‧‧‧ steps of the non-coherent neighbor cell search method of the embodiment of the present invention

FIG. 1A is a schematic diagram showing the structure of signals of PSS and SSS in FDD mode according to a preferred embodiment of the present invention.

FIG. 1B is a schematic diagram showing the signal structure of the PSS and the SSS in the TDD mode according to a preferred embodiment of the present invention.

FIG. 2 is a schematic diagram showing the configuration of signals in the frequency domain and the time domain in the FDD mode according to a preferred embodiment of the present invention.

FIG. 3 is a schematic diagram showing the correspondence between a section identification code and a root index according to a preferred embodiment of the present invention.

FIG. 4 is a schematic diagram showing the position of the SSS and the subcarriers according to a preferred embodiment of the present invention.

FIG. 5 is a schematic diagram showing the correspondence between cyclic shift and group segment identification code according to a preferred embodiment of the present invention.

FIG. 6A is a waveform diagram of Q _k (34, 0, 25, 0) according to a preferred embodiment of the present invention.

FIG. 6B is a waveform diagram of Q _k (34, 0, 29, 0) according to a preferred embodiment of the present invention.

FIG. 7A illustrates a first preferred embodiment of the sequence of embodiment Q _k (29,1,25,0) sequence _{Q k (29,1, u, ε} ) is a schematic view of the present disclosure correlation.

FIG 7B illustrates a second preferred embodiment embodiment of the sequence Q _k (34, -1,25,0) sequence _{Q k (34, -1, u} , ε) is a schematic view of the present disclosure correlation.

FIG. 7C illustrates a second preferred embodiment embodiment of the sequence Q _k (25,0,29, -1) sequences with _{Q k (25,0, u, ε} ) is a schematic view of the present disclosure correlation.

FIG 7D illustrates a first (25,0,34,1) sequence _{Q k (25,0, u, ε} ) associated with a schematic diagram of a preferred embodiment of the sequence Q _k embodiment of the present invention.

8A-8D are respectively shown as a preferred embodiment of the present invention And a correlation diagram of the different combinations of ( ε _s , ε _n ).

FIG. 9 is a flow chart showing a non-coherent neighbor cell search method according to a preferred embodiment of the present invention.

For convenience of description of the embodiments of the present invention, the following communication systems are all exemplified by the 3rd Generation Partnership Project Long-Range Evolution (3GPP-LTE) mobile communication system, but the present invention is not limited thereto.

In the uplink and downlink transmission of 3GPP-LTE, signals are organized into radio frames. Moreover, two different frame structures can support the frequency division duplex (FDD) mode and the time division duplex (TDD) mode, respectively. The radio frame structure is 10 milliseconds and is divided into 10 subframes. Each sub-frame is further divided into two 0.5 millisecond time slots (Time Slot). Each radio frame contains 20 time slots, wherein each time slot contains 7 or 6 orthogonal frequency division multiplex symbols (OFDM Symbol) according to the normal cyclic-prefix mode or the extended cyclic-prefix mode.

In order to obtain the necessary information for the initial cell search and synchronization, two synchronization signals are inserted in the above-mentioned frame structure, which are the primary synchronization signal PSS and the slave synchronization signal SSS, respectively. The main synchronization signals PSS and SSS structures in the FDD mode in the time domain are shown in FIG. 1A, and the PSS and the slave synchronization signals SSS structures in the TDD mode are shown in FIG. 1B.

In the 3GPP-LTE system, the synchronization signal is cycled Sexual transmission, two synchronization signals are transmitted in each 10ms wireless frame. Please refer to FIG. 1A first. In the FDD mode, the PSS is transmitted in the last symbol of the first and eleventh time slots in each radio frame, and the PSS is immediately followed by the symbol of the SSS. . Referring to FIG. 1B, in the TDD mode, the PSS is transmitted in the third symbol of the third and thirteenth time slots, and the SSS is located in the first three symbols of the PSS.

Supported by 3GPP-LTE downlink link system Different transmission bands from 1.4 MHz to 20 MHz, wherein the length of the fast Fourier Transform can vary from 128 to 2048. For all transmission configurations, the synchronization channel used to transmit the synchronization signal is arranged at the center of the 1.25 MHz (DC sub-carrier). The synchronization channel occupies a total of 72 subcarriers, of which only the middle 62 subcarriers are used to transmit PSS and SSS, while the 5 subcarriers on both sides are reserved as empty subcarriers. Taking the FDD mode as an example, the configuration of the signals in the frequency domain and the time domain is as shown in FIG. 2 . Among them, the horizontal axis represents time and the vertical axis represents frequency.

In the 3GPP-LTE system, the cell identification code is defined as ,among them, For the group ID, {0,1,...,167}, For the section identification code, {0,1,2}. Wherein, the broadcasted PSS carries the section identification code Information, while SSS carries a group ID Information.

The frequency domain signal of the main synchronization signal PSS is generated by the Zadoff-Chu sequence. The sequence has good autocorrelation characteristics, and has the most obvious peak value in the right direction. The zero timing is zero, and the cross correlation is in It is zero in a certain range, so the detection performance of the sequence can be improved.

The main synchronizing signal PSS produces a mathematical expression such as the following formula (1), which generates a total of 63 signals, and the value placed at the center subcarrier is 0.

Where u is the root index of the Zadoff-Chu sequence, the value of which is determined by the section identifier Decide, the corresponding relationship is shown in Figure 3. As can be seen from the table of Figure 3, different section identification codes A different root index is generated to select a Zadoff-Chu sequence for transmission as information within the PSS.

The slave sync signal SSS is a sequence of length 62 and consists of two maximum length (ML) sequences ( versus , j = {0, 1}) and the scrambling sequence c _j ( n ) is generated. The mathematical expression of the SSS signal is as follows:

Where n =0, 1, ..., 30. The SSS signals are constructed in an interleaved manner. The sequences D _{i , 2 n} ( m ₀ , m ₁ ) and D _{i , 2 n +1} ( m ₀ , m ₁ ) generated by the above equations (2) and (3) will be modulated by BPSK, and The even and odd subcarriers are placed separately, as shown in FIG. Further, in the above formulas (2) and (3), i is a slot index, and in the TDD mode, i = 1, 11. In FDD mode, i = 0, 10.

The above formulas (2) and (3), versus The maximum length sequence of the periodic shift produced by the shift amount m _j . Cyclic shift ( m ₀ , m ₁ ) and group ID There is a one-to-one correspondence between them, as shown in FIG. In addition, the maximum length sequence There is a one-to-one correspondence with m _j , and the maximum length sequence There is also a one-to-one correspondence with m _j . In addition, the scrambling sequence c _j ( n ) is determined by the segment identifier Determined.

It can be seen from the description of the PSS and the SSS that the cell search performed by the user equipment needs to first decode the segment identification code from the PSS in the received signal. . Then, the receiving end detects the maximum length sequence in the SSS. versus When it is composed of the cyclic shift m _j , the group identification code can be found by the cyclic shift m _j Cell identification code .

[First Embodiment]

For the user equipment in the mobile, in order to perform a fast and effective handover, the user equipment needs to search for the service cell (hereinafter referred to as the current cell) of the current location, and periodically Search for neighboring cells. The present invention proposes a non-coherent method for searching for neighbor cell identification codes, in order to parse out the carried cell identity code from the synchronization signal received by the user equipment, and to distinguish the current The cell identifier of the serving cell and the cell identifier of the neighboring base station help the user device to move when Quick hand change function.

In the first embodiment, how to search for the segment identification code of the current cell and the neighboring cell from the PSS will be explained. For convenience of description of the embodiments of the present invention, it is assumed that the user equipment has obtained the correct symbol timing and the fractional carrier frequency offset from the PSS signal. Moreover, it is assumed that the user equipment has obtained the component on each subcarrier in the PSS signal through the fast Fourier transform, and therefore, the received PSS signal can be expressed as follows:

Where k is a subcarrier index. A frequency domain signal Z _k from the current component included PSS cell element and the components of the adjacent cell element PSS. among them, as well as The channel responses of the current cell and the neighboring cell are respectively ε _s and ε _n are the integer carrier frequency offset (ICFO) of the current cell and the neighboring cell, respectively. as well as The PSS sequences generated according to the root index u _s and u _n are respectively. Where u _s is the root index corresponding to the segment identification code of the current cell, and u _n is the root index corresponding to the segment identification code of the neighboring cell. V _k is an additive white Gaussian Noise (AWGN) with a variation of σ ² , It is a signal-to-interference power ratio.

Since the current cell signal and the signal from the neighboring cell are from different base stations and have gone through different paths, the channel response of the current cell in the above (4) is Channel response with neighboring cells Different, and the frequency drift ε _{s of the} main cells and the frequency drift ε _{n of} adjacent cells may also be different. In this embodiment, the user equipment is within the service range of the current cell and is close to a neighboring cell. In the above case, since the synchronization signal energy emitted by the base station of the current cell is strong in the received signal, and the synchronization signal energy emitted by the base station of the neighboring cell is weak, the frequency domain receiving signal Z _{k is used} for searching. At the time of the segment identification code of the cell, the base station signal energy of the neighboring cell can be regarded as a kind of interference.

For the segment identification code of the current cell, in this embodiment, the frequency domain receiving signal Z _k is detected by using a non-coherent manner, that is, the channel response of the channel is not required to be analyzed. Taking the partial correlation as an example, the frequency domain receiving signal Z _k is directly detected, and the integer frequency drift estimation is performed at the same time. The segment correlation method can be expressed by mathematical expression as follows:

Where G is expressed as the number of segments and C is expressed as the length of each segment. CG = N and N is the number of subcarriers with a value of 63. According to the above formula (5), the root index corresponding to the maximum correlation value will be found. And integer carrier frequency drift value . Where the root index The corresponding segment identification code is the estimated segment identification code of the current cell. In this embodiment, although the segment identification code of the current cell is detected by the segment correlation operation, those skilled in the art should know that the present embodiment can also use a matched filter and differential correlation ( Differential correlation), etc., or use the same tone method to estimate the channel identifier of the current cell after channel estimation of the received signal.

After estimating the segment identification code of the current cell and its integer carrier frequency drift, the present embodiment will perform a segment identification code for searching for neighboring cells. In the receiving signal of the user equipment, the signal energy emitted by the neighboring cell base station is smaller than the signal energy emitted by the current cell base station. Therefore, when detecting the identification code of the neighboring cell, the current cell base station sends out Signals cannot be ignored and cannot be considered interference. The embodiment of the present invention provides a mathematical operation for parsing and simplifying the frequency domain receiving signal Z _k , and multiplying the k-th and k +1 frequency-domain receiving signals Z _k and Z _{k +1} by a known local synchronization signal. (that is, the root index corresponding to the detected segment identifier of the current cell Integer carrier frequency drift with current cells Generated sync signal versus ), then calculate the difference between the two products and generate a synthesized received signal F _k . Through the above operation, the signal component of the current cell base station will be eliminated, and the signal of the adjacent cell base station will be left. Its mathematical expression is as follows:

In the above formula (6), it has been assumed that the current cell is detected. versus To be correct, that is , .

The coherent bandwidth of a general wireless transmission channel is usually much larger than two subcarrier widths. In this embodiment, in order to simplify the above formula (6), it is assumed that channels of adjacent subcarriers ( k and k +1) are assumed. The difference is small and can be ignored, that is to say And . Therefore, the above formula (6) can be simplified as:

It can be seen from the above formulas (6) and (7) that the current cell signal component in Z _k ride on After, only get Similarly, the current cell signal component in Z _{k +1} ride on After that, it can also be reduced to . Moreover, the applicant assumes that the channel effects of adjacent subcarriers are similar, that is, the assumption To make two products versus After the subtraction, the current cell signal components and channel effects in Z _k and Z _{k +1} will be eliminated. In other words, among the actual synthesized signals F _k , the received signal components from the current cell base station Almost completely eliminated, leaving only unknown sync signal components of neighboring cell base stations. Therefore, when detecting adjacent cells in the present embodiment, instead of using the received signal Z _k , the synthesized signal F _k that eliminates the main cell signal component is used.

The synthesized signal F _k in the above formula (7) can be simplified as follows:

In general, the identification code is detected by calculating the magnitude of the correlation. The previous method is to calculate the similarity between the received signal and each possible primary synchronization signal. When detecting the identification code of the neighboring cell, the main synchronization signal is , u _n =25,29,34. However, in the detection of the neighbor cell identification code, the present embodiment does not use the received signal Z _k but uses the synthesized signal F _k that eliminates the current cell signal component. Therefore, if the correlation is used for detection, if the original primary synchronization signal is used For correlation detection, the error rate will increase and the results obtained may be distorted. In theory, when calculating the correlation, the local signal should use the original signal that is not interfered by any channel. As seen from the simplification of equation (8), It can be regarded as a channel-dependent coefficient, therefore, in this embodiment, the applicant uses the correlation between the calculation F _k and Q _k ( u _s , ε _s , u _n , ε _n ) to detect neighboring cells. The section identifier of the element. In other words, will be detected It is represented by a new reference sequence Q _k ( u _s , ε _s , u _n , ε _n ). This sequence _{Q k (u s, ε s} , u n, ε n) is u _s and ε _s is the element corresponding to the cell current root index and the frequency drift. When detecting the identification code of the neighboring cell, both u _s and ε _s are known values.

In addition, the applicant uses the correlation between calculating F _k and Q _k ( u _s , ε _s , u _n , ε _n ) to detect the segment identification code of the neighboring cell, and the reason is that the present invention The embodiment emphasizes the non-coherent neighbor cell search. In other words, when performing the identity code search of the neighbor cell, the channel estimation of the neighbor cell base station is not performed first, and in the above mathematical derivation, see, The correlation with the correlation operation is extremely low. Therefore, in this example, the correlation operation uses the reference sequence Q _k ( u _s , ε _s , u _n , ε _n ) to reduce the receiver's need for channel estimation. The amount of calculations measured.

The above new sequence Q _k ( u _s , ε _s , u _n , ε _n ) is calculated as follows:

Here, in the above formula (9), ε _n = ε _s - Δ. The sequence Q _k ( u _s , ε _s , u _n , ε _n ) has a sinusoidal-like amplitude characteristic, and its waveform diagram is shown in FIGS. 6A and 6B. Of FIG. 6A and 6B are schematic waveform diagram illustrating Q _k (34,0,25,0) and Q _k (34,0,29,0) a. Wherein, the abscissa is the subcarrier k index, and the ordinate is the amplitude. As can be seen from the above formula (1), the component of the center subcarrier of the main synchronizing signal PSS is zero, so that in the waveform diagrams of the 6A and 6B, a recess is formed in the vicinity of the center subcarrier ( k = 30). The sequence Q _k ( u _s , ε _s , u _n , ε _n ) has the same periodicity as the Zadoff-Chu sequence at the rest of the position.

In order to verify that the sequence Q _k ( u _s , ε _s , u _n , ε _n ) is used to detect the detection characteristics of neighboring cell identifiers, the applicant uses the segment correlation algorithm to verify the sequence Q _k ( u _s , ε _s , u _n , ε _n ) autocorrelation and cross-correlation. The first picture shows 7A sequence Q _k (29,1,25,0) sequence _{Q k (29,1, u, ε} ) correlation FIG. The first picture shows 7B sequence Q _k (34, -1,25,0) sequence _{Q k (34, -1, u} , ε) correlation FIG. Figure 7C is a graphical representation of the correlation of the sequence Q _k (25,0,29,-1) with the sequence Q _k (25,0, u , ε ). Figure 7D is a graphical representation of the correlation of the sequence Q _k (25,0,34,1) with the sequence Q _k (25,0, u , ε ). Wherein, the abscissa is an integer carrier drift ε , and the ordinate is the calculated correlation value. As can be seen from the above 7A-7D, the sequence Q _k ( u _s , ε _s , u _n , ε _n ) has a distinct peak at the time of the alignment, and the sequence generated by the different root index u _n The cross-correlation is very low, and thus the sequence Q _k ( u _s , ε _s , u _n , ε _n ) proposed in the present embodiment is verified to maintain excellent detection characteristics.

In order to slow down the channel effect For the distortion caused by the search, this embodiment proposes to use a non-coherent correlation operation to search for the segment identification code of the neighboring cell. By calculating the correlation value of Q _k ( u _s , ε _s , u _n , ε _n ) generated by the synthesized signal F _k for all possible u _n , and finding the u _n having the largest correlation. The correlation operation of this embodiment takes a partial correlation (Partial Correlation) as an example, and its mathematical expression is expressed as:

Since the correspondence between the segment identification code and the root index is one-to-one, the root index with the largest correlation is found by the above formula. After that, the section identification code of the neighboring cell can be found. In addition, in the above (10), the integer frequency drift ε _n estimation of the neighboring cells is also performed simultaneously, however, the influence of the component accuracy of the user equipment and the Buhler effect which may be faced in practice may be considered. In this embodiment, when the integer frequency drift ε _n is estimated, the integer frequency drift difference Δ between the current cell and the adjacent cell is set to a fixed range, for example, Δ={-1, 0, 1}. However, those skilled in the art should know that the range of the integer frequency drift estimation can be adaptively adjusted according to the channel environment of the actual channel or the computing power of the user equipment, etc., therefore, this embodiment does not use this range as limited.

In the above formula (10), G is expressed as the number of segments, and C is expressed as the length of each segment. In a preferred embodiment of the invention, CG = N and C, G are factors of N. N is the number of subcarriers and its value is 63. Due to the different phase changes of the channels in the 63 subcarriers in the wireless channel environment, each product value of 63 points is caused. The phase variation, if a simple correlation operation is used, directly accumulates all the product values, so that the last calculated correlation value is seriously affected by the channel variation. Therefore, in this embodiment, segment correlation is used to segment the correlation operation, and the 63-point signal is divided into a plurality of segments. In each segment, the phase of the channel effect can be regarded as an approximation, so that the last calculated correlation value is less Affected by channel phase changes.

In addition, although the above CG = 63, when calculating the correlation, each point signal on the subcarrier is considered. However, those skilled in the art should know that when calculating the correlation described above, only the signals of a part of the subcarriers can be considered, so that the above CG <63, for example, C = 10, G = 6, and CG = 60. In other words, when the correlation is calculated in this embodiment, the calculated number of signal points can be adaptively adjusted. In addition, in another preferred embodiment, when the accumulated correlation value is greater than a threshold during the operation, it indicates that the root index u _n tested at this time has a great possibility that the composite signal F _{k is} most likely It Therefore, you can stop calculating the correlation value and reduce the amount of unnecessary calculations.

In addition, the section identifier of the neighboring cell in this embodiment is searched by the segment correlation. However, those skilled in the art should know that the correlation operation of the embodiment may also use a matched filter and differential correlation, and the like. Therefore, the invention is not limited thereto.

In this embodiment, the frequency drift is added while searching for the identification code of the current cell. Estimate, and also add frequency drift while searching for identification codes with neighboring cells Estimate. However, those skilled in the art should understand that in other embodiments, the frequency drift can also be ignored, or the receiver has estimated the complete frequency drift before performing the identification code search of this embodiment. The present invention is not limited to estimating the frequency drift while performing the identification code search. In a preferred embodiment of the present invention, if the frequency drift ε _n and ε _{s of the} current cell and the neighboring cell are ignored, the above sequence Q _k ( u _s , ε _s , u _n , ε _n ) can be simplified to Q _k ( u _s , u _n ), the value is . The composite signal F _k can be simplified to The mathematical operation related to the segment in the above formula (10) can be expressed as

[Second embodiment]

In this embodiment, how to search for the group identifier of the current cell and the neighboring cell from the SSS will be explained. Since the SSS signals D _{i , 2 n} ( m ₀ , m ₁ ) and D _{i , 2 n +1} ( m ₀ , m ₁ ) emitted by the base station, the scrambling sequences c ₀ ( n ) and c ₁ ( n ) are included. And the scrambling sequence is determined by the section identification code in the PSS signal. Therefore, in this embodiment, it is assumed that the user equipment has detected the segment identification code of the current cell and the neighboring cell, and assumes that the integer frequency drift ε _s and ε of the current cell and the neighboring cell have been detected. _n . The received SSS signal can be expressed as follows:

Where k is a subcarrier index and i is a slot index. In the received signal, the base station of the current cell has a stronger sync signal energy, and the base station of the neighboring cell transmits weaker sync signal energy. Therefore, when searching for the group identifier of the current cell in this embodiment, The base station signal energy of the neighboring cell can be regarded as a kind of interference, and directly detects the received SSS signal to decode the cyclic shift corresponding to the group identifier of the current cell. . In this embodiment, the segment correlation operation is taken as an example, and it is assumed that the user equipment has not performed frame timing synchronization. Therefore, the received SSS signal is detected and the frame timing is estimated at the same time. The mathematical expression is as follows:

among them,[. ] represents the remainder of 20, where [ i +10] represents the remainder of dividing i +10 by 20, and [ t +10] represents the remainder of dividing t +10 by 20. In this embodiment, it is assumed that when the group identification code is searched, the user equipment has not performed the frame timing synchronization, and it is impossible to determine which time slot the received SSS signal belongs to, so that the optimal frame timing and group identification are obtained. The code, when searching for the above formula (12), needs to be detected for 168 × 2 possible sequences.

In this embodiment, although the group identification code of the current cell is detected by using the segment correlation operation, those skilled in the art should know that the present embodiment can also use a matched filter and differential correlation ( Differential correlation), etc., or use the homology method to estimate the channel identification code of the current cell after channel estimation of the received signal.

Next, this embodiment will detect the group identification code of the neighboring cell. In the receiving signal of the user equipment, the signal energy emitted by the neighboring cell base station is smaller than the signal energy emitted by the current cell base station. Therefore, when detecting the identification code of the neighboring cell, the current cell base station sends out Signals cannot be ignored and cannot be considered interference. In this embodiment, we use the same method as the first embodiment described above to eliminate the signal component of the current cell in the received signal by a simple mathematical operation, and generate a composite signal N _{i , k} , the mathematical expression is as follows :

Since the neighboring cell identification code is detected by the present embodiment, instead of using the received signal Z _{i , k} , the synthesized signal N _{i , k for} canceling the main cell signal component is used, and the derivation of the above formula (13) is used. It can be seen that when the correlation code is used to detect the identification code, the local signal is no longer , but a new detection sequence . among them, .

Similarly, we have to verify the new sequence Whether it has good detection performance. Here, we use the segment-related operations to verify, and the mathematical expressions are as follows:

According to the above formula, we need to target each different And the combination of ( ε _s , ε _n ) to verify the corresponding sequence 168 different sequences with all a and b values Relevant value. Figures 8A~8D are shown separately And a correlation diagram of the different combinations of ( ε _s , ε _n ). Among them, the abscissa is 168 kinds of sequences generated for all a and b values, and the ordinate is the calculated correlation value. Due to applicant verification The amount of data is too large, and only four verification results are randomly presented here. According to the verification results of the above 8A~8D diagrams, it can be seen that only the values of a and b correspond to the correct group identification code. When the peak value appears, the correlation values calculated by the remaining a and b values are quite low. Therefore, the sequence proposed in this embodiment Has excellent detection performance.

After obtaining the synthesized signal N _{i , k for} eliminating the current cell signal component, the embodiment detects the synthesized signal N _{i , k} by the correlation operation to detect the group identification code of the neighboring cell. The measured value 相关( a , b ) of the correlation operation is expressed as:

Next, in the process of searching, calculate the corresponding correlation value for all possible a and b values, and find the corresponding cyclic shift with the largest correlation value. The mathematical expression is as follows:

The frame timing has been estimated due to the aforementioned search for the group identification code of the current cell. Therefore, when searching for the identification code of neighboring cells, only 168 sequences generated by all possible a , b values need to be detected. Since the cyclic shift and the group identification code have a one-to-one correspondence, it is found that the corresponding value has the largest correlation value. , that is, the group identifier of the neighboring cell is searched.

In the above formula (14), G is expressed as the number of segments, and C is expressed as the length of each segment. CG = N and N is the number of subcarriers with a value of 63.

Although this embodiment sets CG = 63, each point on the subcarrier is considered when calculating the correlation. However, those skilled in the art should know that when calculating the correlation described above, only the signals of a part of the subcarriers can be considered, so that the above CG <63, for example, C = 10, G = 6, and CG = 60. In other words, when the correlation is calculated in this embodiment, the calculated number of signal points can be adaptively adjusted. In addition, in another preferred embodiment, when the accumulated correlation value is greater than a threshold in the middle of the operation, it indicates that the cyclic shift a , b , which is tested at this time, is extremely likely to be the synthesized signal N _{t . k is} most likely Therefore, you can stop calculating the correlation value and reduce the amount of unnecessary calculations.

In addition, the group identification code of the neighboring cell in the embodiment is searched by the correlation operation, and the correlation operation is taken as an example of the segment correlation operation. However, those skilled in the art should know that the correlation in this embodiment is relevant. The performance calculation can also use the matched filter and the differential correlation, etc., or can not combine the correlation operation results of the two sets of time slots, and separately check the correlation measurement value of the adjacent cell group identification code of the single time slot, so the present invention does not Limited to this.

It can be seen from the foregoing first and second embodiments that the non-coherent neighbor cell search method proposed by the present invention is simultaneously applicable to searching for the segment identification code of the neighboring cell and the group identification code. In the above two embodiments, the segment identification code and the group identification code of the neighboring cells are obtained by computing a composite signal and then performing a correlation operation. However, those skilled in the art should know that the segment identification code of the neighboring cell and the group identification code may have only one of the identification codes transmitted through the non-coherent search method proposed by the present invention, and the other identification code uses the homology of the prior art. Search or non-coherent search, therefore, the invention is not limited thereto.

The first and second embodiments described above can be summarized into a non-coherent neighbor cell search method, and the method can be at least applied to search for the segment identification code in the PSS and the group identification code in the SSS. This non-coherent neighbor cell search method includes the following steps.

S910: Start identification code search.

S920: Decode the identifier of the current cell. In the above-mentioned first embodiment, the received signal Z _{k is} directly detected by the segment correlation to search for the root index corresponding to the segment identifier of the current cell. . In addition, in the foregoing second embodiment, the received signal Z _{i , k is} directly detected through the fragment correlation to search for a cyclic shift corresponding to the group identifier of the current cell. .

S930: Capture a component on a kth subcarrier and a k +1th subcarrier in a received signal. In the first embodiment, the component of the k-th subcarrier is expressed as the Z _k, a k + 1 sub-carrier component is expressed as Z _{k +1.} In the second embodiment described above, the components of the kth subcarrier are represented as Z _{i , k} , and the components of the k +1th subcarrier are represented as Z _{i , k +1} .

S940: Multiply a component of the k +1th subcarrier in the received signal by a conjugate of components of the k +1th subcarrier in a local synchronization signal to obtain a first product. The local synchronization signal is a synchronization signal corresponding to the identifier of the current cell. In the first embodiment described above, the local synchronization signal is expressed as And the first product is expressed as . In the second embodiment described above, the local synchronization signal is expressed as And the first product is expressed as .

S950: Multiply the component of the kth subcarrier in the received signal by the conjugate of the component of the kth subcarrier in the local synchronization signal to obtain a second product. In the first embodiment described above, the local synchronization signal is expressed as And the second product is expressed as . In the second embodiment described above, the local synchronization signal is expressed as And the second product is expressed as .

S960: Calculate a difference between the first product and the second product to obtain a composite signal. In the first embodiment, the expression of the synthesized signal is expressed as . In the second embodiment, the synthesized signal is represented as N _{i , k} , and the expression is expressed as . It can be seen from the above embodiment that the purpose of the difference is to eliminate the signal component of the current serving cell. Therefore, in the above embodiment, the second product is subtracted from the first product as the synthesized signal, but the technical field has the usual The knowledge learner should know that the result of the difference obtained by subtracting the first product from the second product can achieve the same effect as the above embodiment. Therefore, the present invention does not limit the first product minus the second product as a composite signal.

S970: Perform a correlation operation on all possible identifiers of the neighboring cells by the synthesized signal to obtain an identifier of the neighboring cell. In an embodiment of the invention, a reference sequence function, denoted Y ( x ), is provided as a reference function for the correlation operation. This reference sequence function Y has a correspondence with all possible identification codes of neighboring cells, where x is a possible identification code of the neighboring cells.

In the above-described first embodiment, the kth subcarrier component of the reference sequence function Y is , for example, Q _k ( u _s , ε _s , u , ε ). The correlation operation in the above step S970 is to substitute all possible root indices u of the neighboring cells and the frequency drift ε into the new sequence Q _k ( u _s , ε _s , u , ε ), according to the above ( 10) Calculate the correlation value and find the root index corresponding to the largest correlation value To find the segment identifier of the neighboring cell. In the second embodiment described above, the kth subcarrier component of the reference sequence function Y is , for example, . The correlation operation of the above step S970 is to substitute all possible cyclic shifts a and b values of adjacent cells into a new sequence. After that, the correlation value is calculated according to the above formula (14), and the cyclic shift corresponding to the largest correlation value is found. To find the group ID of the neighboring cell.

S980: End the search.

In the first embodiment and the second embodiment, the method for searching for neighboring cells is to use a non-coherent manner, so that the user equipment can avoid channel estimation of neighboring cells and reduce the decoding of the user equipment. Computation. However, the proposed search method of the present invention can also be applied to user equipment that performs channel estimation of neighboring cells. For example, if the user equipment estimates the channel response of the neighboring cell, the channel response can be first used to eliminate the channel effect in the received synthesized signal, and then the searching method of the present invention is performed. Or, using the estimated channel response, a new sequence function is generated ( or ), and then use this new sequence function to perform correlation operations. Among them, the above The expression is, for example, , The expression is, for example, . In other words, the neighbor cell search method proposed by the present invention can be applied to both homology or non-coherent search of neighboring cells.

In summary, the embodiments of the present invention have at least the following advantages:

1. After the applicant performs a mathematical analysis on the received signal, it proposes a composite signal, which eliminates the signal component and channel effect of the current cell base station in the received signal, leaving the unknown signal component of the adjacent cell base station. Moreover, the synthetic signal is used to perform the identification code search of the neighboring cells, thereby improving the correctness of searching for neighboring cells.

2. A preferred embodiment of the present invention searches for a segment identification code of a neighboring cell, and proposes a new sequence Q _k ( u _s , ε _s , u , ε ) as a reference sequence for performing correlation operations. The autocorrelation peaks of this sequence Q _k ( u _s , ε _s , u , ε ) are obvious, and the cross-correlation values are very low. Therefore, it has very good detection performance, which reduces the detection error rate and slows the channel effect. The signal is distorted. Similarly, a preferred embodiment of the present invention proposes a new sequence when searching for a group identification code of a neighboring cell. It also verified that it has quite good detection performance.

3. A preferred embodiment of the present invention performs identification code search for neighboring cells through segment correlation to mitigate channel effects and to make the calculated correlation values less affected by channel phase variations.

The specific embodiments of the present invention are intended to be illustrative only and not to limit the invention to the above embodiments, without departing from the spirit of the invention and the following claims. The scope of the invention and the various changes made are within the scope of the invention. Therefore, the scope of the invention is defined by the scope of the appended claims.

S910~S980‧‧‧ steps of the non-coherent neighbor cell search method of the embodiment of the present invention

Claims (17)

A non-coherent neighbor cell search method is applicable to a user equipment, wherein the user equipment communicates through a current home cell, and the non-coherent neighbor cell search method includes Decoding the identification code of the current cell; capturing a component on the kth subcarrier and the k +1th subcarrier in a received signal; multiplying a component of the k +1th subcarrier in the received signal by a local synchronization a conjugate of the components of the k +1th subcarrier in the signal to obtain a first product, wherein the local synchronization signal is a synchronization signal corresponding to the identification code of the current cell; the kth of the received signal Multiplying the components of the subcarriers by the conjugate of the components of the kth subcarrier in the local synchronization signal to obtain a second product; calculating a difference between the first product and the second product to obtain a composite signal; the resultant signal for the adjacent cell element identifiers of all possible correlation calculation to obtain the neighbor cell element identification code, wherein, k is an integer, wherein said adjacent cells the synthesized signal of the all-membered Step energy identification code correlation calculation to obtain the neighbor cell element identification code comprising: based on the calculation of the synthesized signal, the providing a reference sequence of functions, expressed as the Y; the M possible identification code is substituted into the reference a sequence function Y , wherein the xth identification code is substituted into the reference sequence function Y to obtain Y ( x ); performing the correlation value operation to obtain M correlation values, wherein the xth possible identification code corresponds to Y ( x) Correlating with the synthesized signal to obtain an xth correlation value; finding a maximum correlation value among the correlation values; and estimating the identification code corresponding to the maximum correlation value as the neighboring cell Identification code, where x and M are positive integers. The non-coherent neighbor cell search method as described in claim 1, wherein the received signal is a primary synchronization signal in a primary synchronization channel, and the identification code of the current cell is a segment identification code, the proximity The identification code of the cell is the segment identification code. a non-coherent neighbor cell search method as described in claim 2, wherein the kth subcarrier component of the reference sequence function Y is represented as Q _k ( u _s , u ), wherein Where u _s is the root index corresponding to the decoded segment identification code of the current cell, u is the root index corresponding to the possible segment identification code of the neighboring cell, and P _k ( u _s ) is the root index u _s Corresponding kth subcarrier component of the main synchronization signal, P _k ( u ) is the kth subcarrier component of the main synchronization signal corresponding to the root index u , and P _{k +1} ( u _s ) is the main synchronization corresponding to the root index u _s The k +1th subcarrier component of the signal, P _{k +1} ( u ) is the k +1th subcarrier component of the primary synchronization signal corresponding to the root index u . The non-coherent neighbor cell search method as described in claim 3, wherein the received signal is represented as Z _k , the number of subcarriers of the received signal is N , and the kth subcarrier component of the local synchronization signal Expressed as P _k ( u _s ), the composite signal is represented as F _k and its value is: Wherein, the correlation operation is a partial correlation, and the operation formula is: Where G is the number of segments, C is the segment length, CG is less than or equal to N , and G , C, and N are positive integers. A non-coherent neighbor cell search method as recited in claim 4, wherein N = 63, and C and G are factors of 63, respectively, and CG = N. The non-coherent neighbor cell search method as described in claim 2, wherein the step of decoding the current cell identifier comprises: decoding a segment identifier of the current cell; and detecting the current The frequency of the primary sync signal of the cell drifts. The non-coherent neighbor cell search method described in claim 6 , wherein the decoded root index corresponding to the segment identifier of the current cell is represented as u _s , and the detected current cell is detected. The integer carrier frequency drift amount of the main synchronization signal is expressed as ε _s , and the kth subcarrier component of the reference sequence function Y is represented as Q _k ( u _s , ε _s , u , ε ), where Where u is the root index corresponding to the possible segment identification code of the neighboring cell, and ε is the possible integer carrier frequency drift of the primary synchronization signal of the neighboring cell, For the kth subcarrier component of the primary synchronization signal corresponding to the root index u _s , P _{k + ε} ( u ) is the kth subcarrier component of the primary synchronization signal corresponding to the root index u , U _s primary synchronization signal corresponding to the root index of the subcarrier components k +1, k +1 subcarrier components of the primary synchronization signals _{P k} +1+ ε _(u) corresponding to the root index u. The non-coherent neighbor cell search method as described in claim 7, wherein the received signal is represented as Z _k , the number of subcarriers of the received signal is N , and the kth subcarrier component of the local synchronization signal is represented. for The composite signal is expressed as F _k and its value is: Wherein, the correlation operation is a partial correlation, and the operation formula is: Where G is the number of segments, C is the segment length, CG is less than or equal to N , and G , C, and N are positive integers. A non-coherent neighbor cell search method as recited in claim 8 wherein N = 63, and C and G are factors of 63, respectively, and CG = N. The non-coherent neighbor cell search method as described in claim 1, wherein the received signal is a slave synchronization signal in a slave synchronization channel, and the identifier of the current cell is a group identity, the proximity The identification code of the cell is a group identification code. A non-coherent neighbor cell search method as recited in claim 10, wherein the kth subcarrier component of the reference sequence function Y is represented as , whose value is: among them, For the decoded cyclic shift corresponding to the group identifier of the current cell, ( a , b ) is a cyclic shift corresponding to the possible group identifier of the neighboring cell, i is a time slot index, and ε _s is The integer carrier frequency drift of the current cell signal, ε _{n is} the integer carrier frequency drift of the neighbor cell signal, Cyclic shift The kth subcarrier component of the corresponding slave sync signal, The k- th subcarrier component of the dependent synchronization signal corresponding to the cyclic shift ( a , b ). The non-coherent neighbor cell search method as described in claim 11, wherein the received signal is represented as Z _{i , k} , the number of subcarriers of the received signal is N , and the kth subcarrier of the local synchronization signal Component is expressed as , the composite signal is expressed as N _{i , k} , and its value is: Wherein, the correlation operation is a partial correlation, and the operation formula is Where G is the number of segments, C is the segment length, CG is less than or equal to N , and G , C, and N are positive integers. A non-coherent neighbor cell search method as recited in claim 12, wherein N = 63, and C and G are factors of 63, respectively, and CG = N. As described in the non-coherent neighbor cell search method described in claim 1, the correlation operation is a partial correlation, a differential correlation, or a matched filter. As described in the non-coherent neighbor cell search method described in claim 1, the step of decoding the identifier of the current cell includes: Perform a channel estimation; use the channel to estimate the result, and search for the identification code of the current cell. For the non-coherent neighbor cell search method described in claim 1, the step of decoding the identifier of the current cell includes: correlating the received signal with all possible identifiers of the current cell. An operation is performed to obtain an identification code of the current cell. A neighboring cell search method is applicable to a user equipment, and is applicable to both a coherent search and a non-coherent search. The user equipment communicates through a current cell, and the neighbor cell search method includes Decoding the identification code of the current cell; capturing a component on the kth subcarrier and the k +1th subcarrier in a received signal; multiplying a component of the k +1th subcarrier in the received signal by a local synchronization a conjugate of the components of the k +1th subcarrier in the signal to obtain a first product, wherein the local synchronization signal is a synchronization signal corresponding to the identification code of the current cell; the kth of the received signal Multiplying the components of the subcarriers by the conjugate of the components of the kth subcarrier in the local synchronizing signal to obtain a second product; calculating a difference between the first product and the second product to obtain a composite signal; The operation of synthesizing the signal provides a reference sequence function corresponding to the synthesized signal, wherein the reference sequence function is a function of the identification code of the neighboring cell; the synthesized signal is the parameter The sequence function performs a correlation operation, wherein all possible identification codes of the neighboring cells are substituted into the reference sequence function, and respectively perform correlation operations on the synthesized signals; and the correlation corresponding to the maximum correlation in the correlation operation The identification code candidate of the cell is the identification code of the neighboring cell, where k is an integer.