Method and system for improving timing position estimation in wireless communications networks

Abstract

A method for determining timing positions in a wireless communications system comprises creating a time-domain timing detection window from a preamble of a receiving signal, generating a first vector of correlations between sampling points in the time-domain timing detection window and sampling points of a known preamble, identifying a pivot position from the largest correlation value of the first vector and generating second vectors based on the pivot position, generating a third vector comprising the largest elements of the second vectors; generating a fourth vector comprising sums of elements in the second vectors, generating fifth and sixth vectors comprising a sum of subsets of the third and fourth vectors, respectively, calculating a seventh vector using the fifth and sixth vectors according to a predetermined equation, and selecting an index of one element from the fifth and seventh vectors to be the timing position according to a predetermined rule.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

The drawings accompanying and forming part of this specification are included to depict certain aspects of the invention. The invention may be better understood by reference to one or more of these drawings in combination with the description presented herein. It should be noted that the features illustrated in the drawings are not necessarily drawn to scale.

FIG. 1 is a flow diagram illustrating a method for obtaining a timing position of a receiving signal according to one embodiment of the present invention.

FIG. 2 illustrates a system that employs the method disclosed in the present invention.

DESCRIPTION

The following detailed description of the invention refers to the accompanying drawings. The description includes exemplary embodiments, not excluding other embodiments, and changes may be made to the embodiments described without departing from the spirit and scope of the invention. The following detailed description does not limit the invention. Instead, the scope of the invention is defined by the appended claims.

The method disclosed in the present invention describes a fine timing estimation algorithm. The method is easy to implement and can significantly improve the accuracy of timing position estimation. Although a WiMax wireless communications network is used to illustrate the system and method of the present invention, one having skills in the art would recognize that the present invention is applicable to the timing estimation process of any wireless communications network that uses unique preambles to facilitate the acquisition of radio signals. Orthogonal frequency division multiple access (OFDMA) and orthogonal frequency division multiplex (OFDM) are examples of such wireless communications networks.

The method disclosed in the present invention first creates a time-domain timing detection window. Correlations between the sampling points in the time-domain timing detection window and a portion of the sampling points of a known preamble are calculated and a vector of correlations is created. The position of the largest value in the vector of correlations is identified. Based on the position information, the time-domain timing detection window is divided into segments.

Two vectors—a sum_res_sum and a sum_res_max—are created from the sampling points in the time-domain timing detection window. A timing position is identified, based on the position of the largest element of the sum_res_sum vector or the sum_res_max vector.

FIG. 1 is a flow diagram illustrating a method for obtaining a timing position of a receiving signal according to one embodiment of the present invention.

In step 110, a preamble is detected in a time-domain sampling window W_{preamble—det}. A coarse estimate of a timing position of the preamble and a carrier frequency offset can be obtained using an algorithm that is known to a person with the skill of the art. The size of the time-domain sampling window W_{preamble—det} equals to the size of the FFT operation. Let L denote the size of the time-domain sampling window W_{preamble—det}.

In step 120, a time-domain timing detection window W_timing—dec dec is created by adding N_inc—range sampling points to both sides of the time-domain sampling window W_{preamble—det}. The size of W_timing—dec is equal to L+2N_inc—range. The value of N_inc—range can be any number, but the typical value of N_inc—range is equal to either one or half of the FFT operation.

Let a vector R represent the sampling points in the time-domain timing detection window W_timing—dec. Let a vector Preamble_TD represent the sampling points of a known preamble. The size of Preamble_TD is L+N_CP, where L is the size of the FFT operation and N_CP is the size of the cyclic prefix (CP).

In step 130, the correlations between a subset of the sampling points in the vector R and a subset of the sampling points in the vector Preamble_TD are calculated according to the following equation: s(k)=|R(k:k+N_CP+┌L/3┐−1)(Preamble_TD(1:N_CP+┌L/3┐))^H|. In the above equation, ┌ . . . ┐ is the ceiling operator; ( . . . )^H is the Hermitian transpose operator; and | . . . | is the absolute value operator. Let the vector of the correlations with K elements be denoted as S=[s(1), s(2), . . . , s(K)], where K=L+2N_inc—range−N_CP−┌L/3┌+1. (The vector S is referred to as the first vector in the Claims section.)

In step 140, an element with the largest value in the vector S is identified and let k_max denote the index of the element. A pivot position k₁ is obtained according to the following equation: k₁=mod(k_max,┌L/3┐), where mod( . . . ) operator produces the reminder of k_max divided by ┌L/3┐.

In step 150, A series of 2d+1 correlation values are taken from the vector S to form subsets, denoted as Ŝ(u), where 1≦u≦U and U=floor((K−k₁−1)/┌L/3┐)+1. The method is based on the following rule: Ŝ(u)=[s(k₁+(u−1)┌L/3┐−d), . . . , s(k₁+(u−1)┌L/3┐+d)], where d is a predetermined positive integer value. (The vector Ŝ(u) is referred to as the second vector in the Claims section.)

Let ŝ_max(u) denote the largest element of Ŝ(u). Let ŝ_sum(u) be the sum of all elements of Ŝ(u). Two vectors are created: Ŝ_max=[ŝ_max(1), ŝ_max(2), . . . , ŝ_max(U)] and Ŝ_sum=[ŝ_sum(1), ŝ_sum(2), . . . , ŝ_sum(U)]. (The vector Ŝ_max and the vector Ŝ_sum are referred to as the third factor and the fourth vector in the Claims section.)

In step 160, a vector sum_res_max is generated from Ŝ_max according to the following equation:

sum_—res_max=[ŝ_max(1), . . . , ŝ_max(U−2)]+[ŝ_max(2), . . . , ŝ_max(U−1)]+[ŝ_max(3), . . . , ŝ_max(U)]

In addition, a vector sum_res_sum is generated from Ŝ_sum according to the following equation:

sum_—res_sum=[ŝ_sum(1), . . . , ŝ_sum(U−2)]+[ŝ_sum(2), . . . , ŝ_sum(U−1)]+[ŝ_sum(3), . . . , ŝ_sum(U)]

The length of both vectors in step 160 equals U−2. (The vectors sum_res_max and sum_res_sum are referred to as the fifth vector and the sixth vector in the Claims section.)

Step 170 illustrates how to find a timing position. There are two alternative approaches to achieve the goal. In one approach, the index of the largest element of the vector sum_res_sum is denoted as i_max. Let tmp1 be the larger of the two values, sum_res_sum(i_max−1) and sum_res_sum(i_max+1). Two predetermined coefficients are chosen and denoted as f₁₁ and f₁₂. Temp1 is then compared with f₁₁*sum_res_sum(i_max). If temp1 is less than or equal to f₁₁*sum_res_sum(i_max), then i_max is identified as the timing position. However, if temp1 is greater than f₁₁*sum_res_sum(i_max), then a new vector is generated according to the following equation: sup er_sum=sum_res_sum+f₁₂*sum_res_max. The index of the largest element of the vector sup er_sum is denoted as j_max, and it is identified as the timing position. (The vector sup er_sum is referred to as the seventh vector in the Claims section.)

According to the other approach, the index of the largest element of the vector sum_res_max is denoted as i_max. Let temp2 be the larger of the two values—sum_res_max(i_max−1) and sum_res_max(i_max+1). Two predetermined coefficients are chosen and denoted as f₂₁ and f₂₂. Temp2 is then compared with f₂₁*sum_res_max(i_max). If temp2 is less than or equal to f₂₁*sum_res_max(i_max), then i_max is identified as the timing position. However, if temp2 is greater than f₂₁*sum_res_max(i_max), then a new vector is generated according to the following equation: sup er_sum=sum_res_max+f₂₂*sum_res_sum. The index of the largest element of the vector sup er_sum is denoted as j_max, and it is identified as the timing position.

FIG. 2 illustrates a system exemplifying the method disclosed in the present invention. The system 200 includes five modules: a preamble detection module 210, a timing detection window module 220, a correlation module 230, a sum_res module 240, and a timing position module 250.

The preamble detection module 210 detects the position of a preamble and determines a coarse timing position and a sub-carrier frequency offset. The timing detection window module 220 creates a time-domain timing detection window using a time-domain sampling window. The time-domain timing detection window W_timing—dec is created by adding N_inc—range sampling points to both sides of the time-domain sampling window W_{preamble—det}. The size of the time-domain sampling window W_{preamble—det} equals to the size of the FFT operation, denoted as L. The size of W_timing—dec is equal to L+2N_inc—range. The value of N_inc—range can be any number, but the typical value of N_inc—range is equal to either one or half of the size of the FFT operation.

The correlation module 230 generates a correlation vector comprising of correlations between a subset of the time-domain timing detection window and a subset of the sampling points of a known preamble. A correlation vector with K elements is denoted as S=[s(1), s(2), . . . , s(K)], where K=L+2N_inc—range−N_CP−┌L/3┐+1.

The correlation module 230 identifies the largest value in the vector S. Let k_max denote the index of the element. A pivot position k₁ is obtained according to the following equation: k₁=mod(k_max,┌L/3┐), where mod( . . . ) operator produces the reminder of k_max divided by ┌L/3┐.

Any subset of the vector S is denoted as Ŝ(u), where 1≦u≦U, and U=floor((K−k₁−1)/┌L/3┐)+1. Let d be a predetermined integer, and 2d+1 correlation values are taken from the vector S to form a series of subsets Ŝ(u). The method to form subsets Ŝ(u) is based on the following equation: Ŝ(u)=[s(k₁+(u−1)┌L/3┐−d), . . . , s(k₁+(u−1)┌L/3┐+d)].

The sum_res module 240 creates two vectors, and they are: Ŝ_max=[ŝ_max(1), ŝ_max(2), . . . , ŝ_max(U)] and Ŝ_sum=[ŝ_sum(1), ŝ_sum(2), . . . , ŝ_sum(U)], where ŝ_max(u) denotes the largest element of Ŝ(u) and ŝ(u) is the sum of all elements of Ŝ(u). Subsequently, two vectors—sum_res_sum and sum_res_max—are created from Ŝ_max(u) and Ŝ_sum(u), respectively, where sum_res_max=[ŝ_max(1), . . . , ŝ_max(U−2)]+[ŝ_max(2), . . . , ŝ_max(U−1)]+[ŝ_max(3), . . . , ŝ_max(U)] and sum_res_sum=[ŝ_sum(1), . . . , ŝ_sum(U−2)]+[ŝ_sum(2), . . . , ŝ_sum(U−1)]+[ŝ_sum(3), . . . , ŝ_sum(U)]. The sum_res module also creates a vector sup er_sum according to the following equation: sup er_sum=sum_res_sum+f₁₂*sum_res_max, where f₁₂ is a predetermined value.

The timing position module 250 identifies the largest element of the vectors sum_res_sum and sup er_sum. The index of the largest element of the vector sum_res_sum is denoted as i_max, and the index of the largest element of the vector sup er_sum is denoted as j_max. The timing position module 250 compares the larger of the two values—sum_res_sum(i_max−1) and sum_res_sum(i_max+1)—with f₁₁*sum_res_sum(i_max), where f₁₁, is a predetermined value. If the larger of the two values—sum_res_sum(i_max−1) and sum_res_sum(i_max+1)—is greater than f₁₁*sum_res_sum(i_max), then j_max is the timing position; otherwise, i_max is the timing position.

By adopting the method disclosed in the present invention, the timing error can be narrowed down to a value between −1 and 1 in an additive white Gaussian noise (AWGN) channel or a value between −5 and 5 in an ITU multi-path channel.

The above illustration provides many different embodiments or embodiments for implementing different features of the invention. Specific embodiments of components and processes are described to help clarify the invention. These are, of course, merely embodiments and are not intended to limit the invention from that described in the claims.

Although the invention is illustrated and described herein as embodied in one or more specific examples, it is nevertheless not intended to be limited to the details shown, since various modifications and structural changes may be made therein without departing from the spirit of the invention and within the scope and range of equivalents of the claims. Accordingly, it is appropriate that the appended claims be construed broadly and in a manner consistent with the scope of the invention, as set forth in the following claims.

Claims

1. A method for determining a timing position in a wireless communications system, the method comprising: creating a time-domain timing detection window from a preamble of a receiving signal;generating a first vector of correlations between a first plurality of sampling points in the time-domain timing detection window and a second plurality of sampling points of a known preamble;identifying a pivot position from the largest correlation value of the first vector and generating three or more second vectors based on the pivot position;generating a third vector comprising large elements of the second vectors;generating a fourth vector comprising a plurality of sums of the elements of the second vectors;generating a fifth and sixth vector comprising a sum of subsets of the third and fourth vectors, respectively;calculating a seventh vector using the fifth and sixth vector according to a predetermined equation; andselecting an index of one element from the fifth and seventh vectors to be the timing position according to a predetermined rule.

2. The method of claim 1, wherein creating the time-domain timing detection window further comprises adding a predetermined number of sampling points on both sides of the time-domain sampling window.

3. The method of claim 1, wherein generating the first vector of correlations comprises calculating the following equation: s(k)=|R(k:k+NCP+┌L/3┐−1)(Preamble_TD(1:NCP+┌L/3┐))H|, where a vector R(k:k+NCP+┌L/3┐−1) represents a subset of a predetermined number of sampling points in the time-domain timing detection window; a vector Preamble_TD(1:NCP+┌L/3┐) represents a subset of a predetermined number of sampling points of a known preamble in a time domain sampling window; ( . . . )H is the Hermitian transpose operator; the notation (a:b) stands for a series of elements from the a-th element to the b-th element; and | . . . | is the absolute value operator.

4. The method of claim 1, wherein the pivot position is the remainder of the index of the largest element of the first vector divided by a first predetermined value.

5. The method of claim 1, wherein each of the second vectors, a subset of the first vector, comprises a predetermined number of correlation values taken from the first vector based on the pivot position.

6. The method of claim 1, wherein the fifth vector is the sum of three subsets of the third vector, each of which has two fewer elements than the third vector.

7. The method of claim 1, wherein the sixth vector is the sum of three subsets of the fourth vector, each of which has two fewer elements than the fourth vector.

8. The method of claim 1, wherein the seventh vector is the sum of the product of the first predetermined value and the fifth vector and the product of a second predetermined value and the sixth vector.

9. The method of claim 1, wherein the predetermined rule is to select the index of the largest element of the fifth vector as the timing position if the larger neighboring element of the largest element of the fifth vector is less than or equal to the product of the largest element and a third predetermined value.

10. The method of claim 1, wherein the predetermined rule is to select the index of the largest element of the seventh vector as the timing position if the larger neighboring element of the largest element of the fifth vector is greater than the product of the largest element and the third predetermined value.

11. A system for improving timing position estimation in a wireless communications system, the system comprising: a preamble detection module determining a coarse timing position and a sub-carrier frequency offset;a timing detection window module creating a time-domain timing detection window using a time-domain sampling window;a correlation module generating a first vector of correlations between the sampling points in the time-domain timing detection window and a subset of the sampling points of a known preamble, obtaining a pivot position, and generating three or more second vectors;a sum_res module creating a third and fourth vectors by using three or more second vectors, creating a fifth vector by using the third vector, creating a sixth vector by using the fourth vector, creating a seventh vector by using the fifth and the sixth vectors according to a predetermined equation;a timing position module determining the timing position of a receiving signal.

12. The system of claim 11, wherein the timing detection window module creates the time-domain timing detection window by adding a predetermined number of sampling points on both sides of the time-domain sampling window.

13. The system of claim 11, wherein the correlation module calculates the correlations between the sampling points in the time-domain timing detection window and the sampling points of the known preamble according to the following equation: s(k)=|R(k:k+NCP+┌L/3┐−1)(Preamble_TD(1:NCP+┌L/3┐))H|, where a vector R(k:k+NCP+┌L/3┐−1) represents a subset of a predetermined number of sampling points in the time-domain timing detection window; a vector Preamble_TD(1:NCP+┌L/3┐) represents a subset of a predetermined number of sampling points of the known preamble in a time domain sampling window; ( . . . )H is the Hermitian transpose operator; and | . . . | is the absolute value operator.

14. The system of claim 11, wherein the correlation module identifies the largest correlation value of the correlations between the sampling points in the time-domain timing detection window and the sampling points of the known preamble.

15. The system of claim 11, wherein the correlation module selects a pivot position by obtaining the remainder of the index of the largest correlation value divided by a first predetermined value.

16. The system of claim 11, wherein the correlation module generates three or more second vectors, each of which is a subset of the first vector and consists of a predetermined number of correlation values taken from the first vector based on the pivot position.

17. The system of claim 11, wherein the sum_res module generates a third vector comprising of the largest elements of the three or more second vectors and a fourth vector comprising of the sum of the elements of the three or more second vectors.

18. The system of claim 11, wherein the sum_res module generates a fifth vector by taking the sum of three subsets of the third vector, and a sixth vector by taking the sum of three subsets of the fourth vector.

19. The system of claim 11, wherein the sum_res module generates a seventh vector by adding the product of the first predetermined value and the fifth vector and the product of a second predetermined value and the sixth vector.

20. The system of claim 11, wherein the timing position module determines a timing position by selecting an index of one element from the fifth and the seventh vectors according to a predetermined rule.

21. The system of claim 20, wherein the predetermined rule is to select the index of the largest element of the fifth vector as the timing position if the larger neighboring element of the largest element of the fifth vector is less than or equal to the product of the largest element and the third predetermined value.

22. The system of claim 20, wherein the predetermined rule is to select the index of the largest element of the seventh vector as the timing position if the larger neighboring element of the largest element of the fifth vector is greater than the product of the largest element and the third predetermined value.