WO2009155882A1 - 一种随机接入前导的检测方法 - Google Patents

一种随机接入前导的检测方法 Download PDF

Info

Publication number
WO2009155882A1
WO2009155882A1 PCT/CN2009/072486 CN2009072486W WO2009155882A1 WO 2009155882 A1 WO2009155882 A1 WO 2009155882A1 CN 2009072486 W CN2009072486 W CN 2009072486W WO 2009155882 A1 WO2009155882 A1 WO 2009155882A1
Authority
WO
WIPO (PCT)
Prior art keywords
value
time domain
point
search window
rach
Prior art date
Application number
PCT/CN2009/072486
Other languages
English (en)
French (fr)
Inventor
薛妍
翟羽佳
谭欢喜
Original Assignee
中兴通讯股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2009155882A1 publication Critical patent/WO2009155882A1/zh

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L27/00—Modulated-carrier systems
    • H04L27/26—Systems using multi-frequency codes
    • H04L27/2601—Multicarrier modulation systems
    • H04L27/2647—Arrangements specific to the receiver only
    • H04L27/2655—Synchronisation arrangements
    • H04L27/2662—Symbol synchronisation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L27/00—Modulated-carrier systems
    • H04L27/26—Systems using multi-frequency codes
    • H04L27/2601—Multicarrier modulation systems
    • H04L27/2647—Arrangements specific to the receiver only
    • H04L27/2655—Synchronisation arrangements
    • H04L27/2656—Frame synchronisation, e.g. packet synchronisation, time division duplex [TDD] switching point detection or subframe synchronisation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W74/00—Wireless channel access, e.g. scheduled or random access
    • H04W74/08—Non-scheduled or contention based access, e.g. random access, ALOHA, CSMA [Carrier Sense Multiple Access]
    • H04W74/0866—Non-scheduled or contention based access, e.g. random access, ALOHA, CSMA [Carrier Sense Multiple Access] using a dedicated channel for access

Definitions

  • the present invention relates to an access signal detection technology in the field of mobile communications, and in particular, to a method for detecting a random access preamble.
  • the mobile terminal In the Long Term Evolution (LTE) system, the mobile terminal first performs downlink synchronization through a synchronization channel (SCH, Synchronization Channel), and finds a radio frame, a reception start point of the subframe, and a cell ID (Cell ID);
  • the broadcast channel BCH, Broadcast Channel
  • RACH random access channel
  • the location of the RACH channel is found based on the radio frame and the reception start point of the subframe determined in the following line synchronization, and the starting point of the transmission preamble is determined, and then one of the available sequences is randomly selected.
  • the preamble is sent.
  • the base station detects the preamble to determine the timing adjustment amount of the uplink synchronization, and sends the timing adjustment amount to the mobile terminal.
  • the mobile terminal adjusts the transmission timing of the uplink signal according to the timing adjustment amount to implement time synchronization of the uplink channel.
  • the length of the ZC sequence is specified as 839 in the LTE standard, and each cell (Cell) has 64 sequences for the preamble.
  • the 64 sequences may be different cyclic shifts from the same root sequence, or different cyclic shifts from different root sequences.
  • the ZC sequence is a constant amplitude zero autocorrelation
  • the correlation between different cyclic sequences of the same root sequence is 0; the correlation of different root sequences (including their cyclic shift sequences) is 1 different root sequences of RACH The correlation between them is very small and can be considered to be approximately equal to zero. Therefore, the correlation correction property of the RACH sequence can be used to perform time domain correlation detection on the random access signal to obtain an uplink timing adjustment amount.
  • the method for detecting the time domain correlation is defined as multiplying and summing the received signal and the local sequence cyclically complex conjugate point to obtain a time domain correlation value of each cyclic shift sample, which is mathematically equivalent
  • the received frequency domain signal is multiplied with the local frequency domain sequence complex conjugate point and converted to the time domain by inverse Fourier transform.
  • the frequency domain form is the local sequence time domain form x ( ), and the frequency domain form is the local sequence complex conjugate.
  • the time domain form is x * ( ), the frequency domain form W;
  • the mobile terminal of the RACH sequence with different cyclic shifts of the root sequence converts the received signal into a complex conjugate point multiplication of the frequency domain with the frequency domain sequence of the root sequence, and then multiplies by frequency domain and inverse Fourier Transforming into the time domain, you can get the time domain related information corresponding to each cyclic shift.
  • FIG. 1 is a schematic diagram of a method for detecting RACH time domain correlation in the existing RACH.
  • this equivalent processing method does not need to perform time domain correlation on the local sequence of each RACH, but only for each local The root sequence performs time domain correlation operations. This has the advantage of reducing the amount of correlation on the receiving side and storing only the local root sequence without storing every possible preamble ⁇
  • the discrete Fourier transform (DFT) and the inverse discrete Fourier transform (IDFT) complexity of the prime are 4 ⁇
  • the fast Fourier transform (FFT) is generally used in the form of the power of 2.
  • IFFT inverse fast Fourier transform
  • the correlation values of each root sequence on different cyclic shifts can be obtained.
  • the correlation value exceeds a certain detection threshold, the correlation value is considered to be a result of the correlation operation of the RACH signal, and the time point corresponding to the correlation value is regarded as timing information; otherwise, the correlation value is regarded as the value of the noise.
  • the detection threshold is also called an absolute threshold of peak detection.
  • Figure 2 is a schematic diagram of the data processing process of RACH peak detection. As shown in Figure 2, the correlation detection results of two RACH users using different cyclic shifts of the same root sequence are given. Points A, B, and C in the figure. , D point, E point, F point, G point and H point are related values of the threshold, wherein point B is the real timing point of user 1, and point F is the real timing point of user 2.
  • the receiver uses the technique of time domain correlation in the time domain correlation operation.
  • the use of the sample technology can reduce the complexity of the implementation and improve the resolution of the time domain, it will bring about the relevant value points.
  • Energy dispersion which can cause uncertainty in timing position.
  • There is a RACH signal in the next search window and its correlation value is at the beginning of the search window.
  • the previous search window will result in the previous search window.
  • There is also a peak The appearance of this peak is a false alarm.
  • the signal energy of user 2 is dispersed into the previous search window due to the use of the technique, and the D point also exceeds the detection. Threshold, forming a false alarm.
  • the detection and capture performance of the RACH signal can be characterized by the preamble miss detection rate, false alarm rate, and detection threshold.
  • the miss detection rate of the RACH signal can be tested by determining the corresponding detection threshold.
  • the false alarm rate is defined as the probability of the detected preamble when there is no signal transmission in the time domain correlation detection.
  • the missed detection rate is defined as the probability that when a preamble different from the transmission is detected, or a preamble is not detected at all, or a correct preamble is detected but an erroneous timing estimate occurs.
  • the required target false alarm rate is 10-3 or less, but adjacent to 10-3, reaches a predetermined false alarm rate when no signal is transmitted during the test, but adjacent to 10-3 or less when the 10-3
  • the detection threshold is the threshold for RACH detection performance.
  • the absolute threshold of RACH peak detection is related to the magnitude of the noise power.
  • the threshold is independent of the noise power, so the definition is as described.
  • the detection threshold is the ratio of the signal correlation value to the noise power as a relative threshold.
  • the ratio of the correlation value of the received signal to the noise power is greater than the relative threshold, it is considered that there is a RACH signal.
  • the relative threshold takes a value of 10-3 at the false alarm rate. This relative threshold is when no signal is sent. Measured, so in the RACH preamble detection, the relative threshold has been pre-set.
  • the noise power is obtained by averaging the relevant data without the RACH signal, that is, the noise and the interference data, and therefore the correlation values of those RACH signals need to be removed.
  • the relevant data without the RACH signal that is, the noise and the interference data
  • the correlation values of those RACH signals need to be removed.
  • each time domain correlation value is judged whether it is noise data, the processing complexity is high and the delay is large.
  • the received signal is a signal that passes through multiple propagation paths.
  • the delay and power of the signal on each propagation path to the receiver are also different. This is called multipath effect. Due to the multipath effect, a plurality of different delay correlation values of the same RACH signal appear in a search window. When there are multiple multipath signal correlation value points exceeding the threshold, how to select one at the point exceeding the threshold A reasonable timing position is also uncertain.
  • the mobile terminal performs the uplink synchronization access system through the RACH channel, and the access success rate is low or the false alarm rate is high due to the inability to quickly and accurately estimate the noise power.
  • the complexity of the processing is high, and there are also defects that the false alarm rate caused by the use of the sampling technique is high and the accuracy of the obtained uplink timing adjustment information is insufficient.
  • the main purpose of the present invention is to provide a method for detecting a random access preamble, which provides accurate and reliable uplink timing adjustment information for a mobile terminal, so as to improve reliability of the mobile terminal accessing the mobile communication system, and reduce processing complexity.
  • the signaling overhead caused by the police.
  • a method for detecting a random access preamble comprising:
  • step B Select one of the available search windows corresponding to the selected local root sequence that has not been detected. Using the search window as the current search window, detecting the time domain correlation value in the current search window, obtaining the detection result value, and sending the detection result value and the corresponding time domain position to step C for processing; c. Determining the detection result value and its corresponding time domain position according to a preset selection condition, and determining a timing position of the RACH preamble signal.
  • the method further includes: performing a down sample processing on the received preamble signal, performing FFT processing, extracting a frequency domain RACH signal, and selecting one of the root sequences allocated from the current cell. The local root sequence that was detected.
  • time domain correlation processing in step A is specifically: multiplying the complex conjugate point with the selected local root sequence, and then transforming the signal into the time domain by IFFT processing.
  • the noise average estimation and the absolute threshold calculation for the time domain RACH signal in step A are specifically: searching for the maximum value of all time domain correlation values of each RACH signal root sequence, which is less than the maximum value nl times The correlation value is averaged, and the average value of the temporary noise is multiplied by a preset relative threshold of n2 times to obtain a noise threshold value, and then the correlation value smaller than the noise threshold is averaged to obtain the The noise mean of the sequence. Finally, the product of the noise mean multiplied by the relative threshold is used as an absolute threshold; wherein nl and n2 are both positive numbers less than one.
  • the value of the detection result obtained in step B is specifically: searching for a maximum correlation value in each search window, and when the maximum correlation value exceeds an absolute threshold, each of the search windows exceeds an absolute threshold The point of the value and the correlation values corresponding to the points adjacent to the left and right are averaged, and when the obtained average value still exceeds the absolute threshold, the correlation value exceeding the absolute threshold value is used as the detection result value.
  • the time domain correlation value in the current search window is detected in the step B, if the detection result value is not obtained, there is no RACH signal in the search window.
  • one of the preset screening conditions in step C is: m times of the detection result value is greater than a maximum correlation value of the current search window, where: m is a positive number greater than 1;
  • the second screening condition in the step C is:
  • the third screening condition in the step C is:
  • a position corresponding to a point in the current search window that is shorter than a certain time length of the first point exceeding the absolute threshold and having a maximum correlation value among the result values is used as the timing position.
  • the step C is specifically: determining a time domain position corresponding to the detection result value satisfying the three preset screening conditions as the determined unique timing position; otherwise, determining that there is no RACH signal in the search window.
  • the present invention processes the signals received by the mobile terminal, and replaces the DFT and IDFT processes by FFT and IFFT processing, respectively, and reduces the frequency domain RACH signal in the extracted signal and complex conjugate with the local frequency domain root sequence.
  • the present invention performs noise mean estimation and absolute threshold calculation on the time domain signal after IFFT processing, which reduces the noise interference in the RACH signal, and provides a reliable theoretical basis for the system to further process the RACH signal, and at the same time, is set reasonably.
  • Absolute threshold which reduces the system's missed detection rate and false alarm rate.
  • the present invention finds the maximum correlation value in each search window, and averages the maximum correlation value and several points adjacent to the left and right, so that the system can quickly and accurately find the search window with the RACH signal, and Reduced the occurrence of false alarms.
  • the timing position is determined by the present invention, by setting three screening conditions of the timing position, the false alarm caused by the use of the sampling technique or due to noise is effectively suppressed, thereby reducing the unnecessary letter caused by the false alarm.
  • the timing point can be obtained relatively accurately, eliminating the timing error caused by the excessive sampling; and passing the detection result value time domain exceeding the absolute threshold
  • the limitation of the distance of the position ensures that the timing position is at a position where the energy is large and the timing deviation is small.
  • FIG. 1 is a schematic diagram of a method for detecting RACH time domain correlation in the prior art
  • FIG. 2 is a schematic diagram of a data processing process of RACH peak detection
  • FIG. 3 is a flowchart of a RACH time domain correlation detection process according to an embodiment of the present invention. Preferred embodiment of the invention
  • the bandwidth occupied by the RACH signal in the frequency domain is 1.08 MHz, and the interval between each RACH subcarrier is specified to be 1.25 kHz.
  • the time domain format of the RACH signal is also specified in the LTE standard: the cyclic prefix (CP, Cycle Prefix) is followed by the preamble of the RACH, and the preamble of the RACH is followed by the guard interval. There is no signal sent during the guard interval.
  • FIG. 3 is a flowchart of a RACH time domain correlation detection process according to an embodiment of the present invention. As shown in FIG. 3, the process includes:
  • Step 301 The base station extracts the preamble portion of the received signal, performs a down sample, and performs FFT processing to extract the RACH signal in the frequency domain data.
  • the received signal is sent to the CP, and then dropped from 30.72 MHz to 1.28 MHz, and then 1024 points of FFT processing is performed to extract 839 points of the RACH frequency domain signal;
  • Step 302 Determine whether all the root sequences allocated to the current cell have been detected. If all the root sequences have been detected, go to step 311; otherwise, go to step 303.
  • Step 303 Perform root sequence correlation detection, and select an undetected root sequence from the root sequence allocated by the current cell.
  • the selected root sequence that is, the local frequency domain root sequence.
  • Step 304 Multiply the received frequency domain RACH signal with the selected local frequency domain root sequence complex conjugate point, and then transform the RACH signal into the time domain by IFFT processing to obtain time domain related data.
  • Step 305 The base station performs noise mean estimation and absolute threshold calculation on a time domain RACH signal of the selected local root sequence obtained by RACH time domain correlation processing.
  • the noise mean estimation and the absolute threshold are calculated as: finding the maximum value of all time domain correlation values of each RACH signal root sequence, and searching for data less than the maximum value n (0 ⁇ " ⁇ 1) times For the mean value, the average value obtained is used as the temporary noise mean value, and the temporary noise mean value is multiplied by a relative threshold value of n ( 0 ⁇ " ⁇ 1 ) times to obtain a noise threshold value.
  • the correlation value data smaller than the noise threshold is generally regarded as the noise interference data not including the RACH signal.
  • the mean value of the correlation value data smaller than the noise threshold is the noise power mean of the root sequence, and the noise power mean is simply referred to as the noise mean value.
  • the product of the noise mean and the relative threshold is called the absolute threshold.
  • the maximum point is point B, where n is 3/5
  • the calculation is less than the maximum value of 3/
  • the average value of the relevant data is 5 times, and the average value of the temporary noise is obtained.
  • the temporary noise average is multiplied by a relative threshold of 3/5 times to obtain the noise threshold.
  • the correlation data smaller than the noise threshold value is regarded as the noise interference data not including the RACH signal, and the mean value of the correlation value data smaller than the noise threshold is calculated to obtain the noise mean value of the root sequence.
  • the mean value of noise power that is, the mean value of noise
  • the product of the noise false alarm rate relative to the mean value detection gate 10-3 when the detection limit as the absolute threshold value referred to as an absolute threshold absolute detection threshold.
  • Step 306 It is determined whether all available search windows corresponding to the selected local root sequence have been detected. If all are detected, step 302 is performed to perform detection of the next root sequence; otherwise, step 307 is performed, that is, from the selected An available search window that is not detected is selected from the available search windows corresponding to the local root sequence for detection.
  • Step 307 Select an available search window that has not been detected from the available search windows corresponding to the selected local root sequence, and determine the start and end positions of the selected available search windows that have not been detected.
  • the selected undetected available search window position is determined by: in the LTE standard, the search window length Ncs value used for each root sequence is given, and each root sequence can be calculated.
  • the 64 preamble sequences of the cell are composed of 64 cyclic shift sequences of the second root sequence.
  • the first search window position is from the first sample to the 13th sample. If the 2048-point IFFT is considered, the position of the first search window is From the first sample to the 32nd sample.
  • Step 308 Detecting a signal correlation value in the selected search window that has not been detected, and selecting a maximum correlation value point in the available search window that has not been detected exceeds an absolute threshold, and any one exceeding an absolute threshold When the average value of the correlation values of the point and its neighboring points still exceeds the absolute threshold, step 309 is performed; otherwise, it is considered that there is no signal available in the current search window, and the process returns to step 306.
  • the suspected signal point refers to each point in the search window that exceeds the absolute threshold and several points adjacent to each other on the left and right.
  • the selected signal in the available search window that has not been detected is detected, specifically: searching for the maximum correlation value point in the selected search window that is not detected, when the maximum correlation value exceeds the absolute threshold, The average value of each point in the available search window that exceeds the absolute threshold and the adjacent points of the left and right sides are averaged, and any one of the obtained points exceeding the absolute threshold and its adjacent ones If the average value of the correlation values of the points still exceeds the absolute threshold, step 309 is performed; otherwise, it is considered that each of the selected available search windows that are not detected does not satisfy the condition of performing step 309, that is, the selected one is considered If there is no signal available in the available search window that has not been detected, the current signal detection process is exited, and the process returns to step 306 to start detecting the signal in the next search window.
  • step 309 is performed to perform the determination process of the timing position.
  • the maximum correlation value in the search window does not exceed the absolute threshold; or the maximum correlation value in the search window exceeds the absolute threshold, but each is absolutely absolute The point of the threshold and the average of one of the adjacent points on both sides of the threshold do not exceed the absolute threshold, indicating that there is no RACH signal.
  • the points adjacent to the two sides of the point exceeding the absolute threshold are listed in chronological order of the correlation value points adjacent to the point and adjacent to the point, which may be points exceeding the absolute threshold, It can be a point that does not exceed the absolute threshold.
  • Using the above processing strategy can ensure that the search window with the current RACH is quickly found. As shown in the time domain related data shown in Figure 2, according to the above processing method, it can be quickly determined that the search windows of point A, point B, point C, point D, point E, point F, point G, and point H are located. There may be signals; this avoids the occurrence of false alarms due to the detection of non-signal points caused by the use of the technique, and ensures that the actual value of the actual timing point is not missed.
  • Step 309 Determine the timing position of the transmission preamble starting point according to the set filter condition, and determine whether any of the selected unsearchable available search windows exceeds the absolute threshold, if yes, proceed to step 310; Otherwise, returning to step 306, indicating that the value of each point cannot satisfy the timing position condition, it is also considered that there is no signal available in the current search window, that is, the signal detection process of the current search window is exited, and the next one is not detected. The search window is detected. .
  • the system considers that the timing position is found, otherwise it determines that the point is not the timing position.
  • the three conditions that need to be met at the same time are as follows:
  • Condition one is that m times the point in the available search window that has not been detected that exceeds the absolute threshold is greater than the maximum correlation value of the search window, where: m is a positive number greater than 1; condition 2 is at the end of the search window k
  • the maximum value of the points is smaller than the mean value of the points at which the point exceeding the absolute threshold is adjacent to the two points adjacent to the two sides, or the position of the point exceeding the absolute threshold is not within the range of k points at the end of the search window, where k is a positive integer
  • the third condition is that the time position in the current search window and the time point of the first point exceeding the absolute threshold is less than a certain length of time, and the point corresponding to the point having the largest correlation value among the points exceeding the absolute threshold is the timing position.
  • Condition 1 The value of 3 times the correlation value corresponding to the point exceeding the absolute threshold is greater than the maximum correlation value of the search window;
  • Condition 2 The maximum value among the 5 points in the tail of the search window is smaller than the point exceeding the absolute threshold
  • Adjacent point correlation value The mean value or the position of the point exceeding the absolute threshold is not within 5 points of the end of the search window;
  • Condition 3 The time domain distance between the current search window and the first correlation value point exceeding the absolute threshold is less than a certain length of time, and exceeds The position corresponding to the point where the correlation value is the largest among the detection result values of the absolute threshold is used as the timing position.
  • the predetermined length of time may be set to 2.56 microseconds.
  • the A point, the B point, the C point It may be a timing point, but only the B point that satisfies the timing condition one, and actually only the B point is the real timing point of the user 1.
  • Satisfying the condition 2 can eliminate the false alarm caused by the excessive correlation of the search window of the search window to the adjacent search window, as shown in FIG. 2, because the signal energy of the user 2 is dispersed to an adjacent search window.
  • the previous search window causes the D point to exceed the absolute threshold, but the point is located within 5 points at the end of the current search window, so it can be positively assumed that there is no RACH signal in the adjacent search window, thereby effectively suppressing the processing virtual
  • Satisfying condition three ensures that the timing position is at a position where the energy is large and the timing deviation is small.
  • the timing position is at a position where the energy is large and the timing deviation is small.
  • Step 310 When there are all points in the selected search window that are not detected that exceed the absolute threshold, the timing position corresponding to the three screening conditions is converted into the timing adjustment amount. Provides uplink synchronization time advance information for users who send this preamble.
  • Step 311 After all the preamble sequences allocated by the cell are detected, the RACH preamble detection process of the cell is ended.
  • an effective noise estimation can be performed for each root sequence, and an absolute threshold can be detected according to the estimated noise, and a RACH based on different cyclic shifts of the same sequence can be used. Users are applicable. Because the signal detection threshold is accurate and real-time, the RACH user can be detected quickly and accurately.
  • the detection method provided by the present invention also considers the suppression of false alarms caused by excessive noise or noise, and reduces false alarms. Unnecessary signaling overhead; By averaging several time-domain correlation values that may be timing points, the timing points can be obtained relatively accurately and the occurrence of false alarms can be reduced, and the timing error caused by excessive sampling can be eliminated.
  • the invention processes the signals received by the mobile terminal, and replaces the DFT and IDFT processing processes by FFT and IFFT processing respectively, and reduces the frequency domain RACH signal in the extracted signal and the complex conjugate point multiplication with the local frequency domain root sequence. Calculated computational complexity and data processing difficulty; The noise mean estimation and absolute threshold calculation of the time domain signal after IFFT processing reduces the noise interference in the RACH signal, which provides a reliable theoretical basis for the system to further process the RACH signal.
  • the timing point can be obtained relatively accurately, eliminating the timing error caused by the excessive sampling; and by passing the detection result value exceeding the absolute threshold
  • the limitation of the distance of the time domain position ensures that the timing position is located at a position where the energy is large and the timing deviation is small.
  • the present invention has extremely strong industrial applicability.

Description

一种随机接入前导的检测方法
技术领域
本发明涉及移动通信领域的接入信号检测技术， 尤其涉及一种随机接入 前导的检测方法。
背景技术
在长期演进（ LTE , Long Term Evolution ) 系统中， 移动终端开机之后首 先通过同步信道（SCH, Synchronization Channel )进行下行同步， 找到无线 帧、子帧的接收起点及小区号（Cell ID ) ;然后检测广播信道（BCH, Broadcast Channel ) , 获取系统信息， 该系统信息包含随机接入信道（ RACH, Random Access Channel )的配制信息； 最后通过所述 RACH信道进行上行同步， 完成 接入系统的工作。
在移动终端上行同步的过程中， 首先以下行同步时确定的无线帧及子帧 的接收起点为基础找到 RACH信道的位置， 并确定发送前导的起点， 然后从 可用的序列中随机地选择一条作为前导发送。 基站对前导进行检测， 以确定 上行同步的定时调整量， 并将其发送给移动终端， 移动终端根据所述定时调 整量对上行信号的发送时刻进行调整， 以实现上行信道的时间同步。
现有 LTE系统的上行随机接入前导使用的是 ZC ( Zadoff-Chu )序列， 第 u 个根 ZC序列定义为： (") = e_ N- , 0≤"≤NZC - 1。 其中： 为所述 ZC 序列的长度， 在 LTE标准中规定为 839, 每个小区 （ Cell )有 64条用于前导 的序列。 这里， 所述 64条序列既可以是来自同一个根序列的不同循环移位， 也可以是来自不同根序列的不同循环移位。 所述 ZC序列是恒幅零自相关
如下特点： 相同的根序列的不同循环序列之间的相关性为 0; 不同的根序列 (包括其彼此的循环移位序列） 的相关性是1 即 RACH的不同根序列 之间的相关性非常小， 可以视为近似等于 0。 因此， 可以利用 RACH序列的 相关性质对随机接入信号进行时域相关检测来获得上行的定时调整量。 所述时域相关检测的方法， 定义为对收到的信号和本地序列循环移位复 共轭点乘求和， 得到每个循环移位样点的时域相关值， 在数学上可以等效成 接收的频域信号与本地频域序列复共轭点乘后通过反傅里叶变换转换到时 域。
假设接收到的信号时域形式为 ) , 频域形式为 本地序列时域形 式为 x( ) , 频域形式为 本地序列复共轭的时域形式为 x*( ), 频域形式 W； 两者的相关函数 ∞) , 则用公式表示为： r(m) =∑ x (n)■ y(m + n) = - Ύ Υ (k)■ X k) · e N 因此， 对使用同一个根序列的不同循环移位的 RACH序列的移动终端而 言， 将接收到的信号转换到频域与该根序列的频域序列复共轭点乘， 再通过 频域点乘以及反傅里叶变换到时域， 可以得到每个循环移位对应的时域相关 信息。
图 1为现有 RACH时域相关检测的方法示意图， 如图 1所示， 通过这种 等效的处理方法， 不需要对每个 RACH的本地序列进行时域相关， 而只需对 每个本地根序列进行时域相关运算即可， 这样做的好处是， 既可以减少接收 侧的相关运算量， 并且只需存储本地根序列， 而不用存储每个可能的前导序 歹 |J。 在系统实现时， 由于 RACH序列的长度是 839 , 使用素数的离散傅立叶 变换（DFT )及反离散傅立叶变换（IDFT ) 复杂度 4艮高， 一般使用 2的次幂 的形式用快速傅立叶变换（FFT )和反快速傅立叶变换（IFFT ) 来分别代替 DFT和 IDFT进行处理。 为保证有用信息不发生损失， 所述 2的次幂的值一 定要大于 839 ,这就是说，要在时域相关运算中引入过釆样， 比如，使用 2048 点 IFFT代替 839点的 IFFT, 这相当于进行了约 2.4倍的过釆样。
通过时域相关的运算，可以得到每个根序列在不同循环移位上的相关值。 当所述相关值超过一定的检测门限时， 认为该相关值是 RACH信号进行相关 运算的结果， 该相关值对应的时间点被认为是定时信息； 否则， 该相关值被 认为是噪声的数值。 这里， 所述检测门限也叫峰值检测的绝对门限。 图 2为 RACH峰值检测的数据处理过程示意图， 如图 2所示， 给出了两 个 RACH用户使用相同根序列的不同循环移位时的相关检测结果，图中 A点、 B点、 C点、 D点、 E点、 F点、 G点和 H点为过门限的相关值， 其中， B点 为用户 1真正的定时点， F点为用户 2真正的定时点。
接收机在进行时域相关运算时釆用了过釆样技术， 使用过釆样技术虽然 可以减少实现的复杂度并提高时域的分辨率， 但由于过釆样会带来各个相关 值点的能量分散， 这可能会造成定时位置的不确定性。 譬如： 由于釆用了过 釆样技术， 用户 1的搜索窗中， 有八、 B和 C三点超过检测门限， 而实际上 只有 B点才是定时点。 使用过釆样在某些情况还会引起一些虚警， 比如： 后 一个搜索窗中有 RACH信号， 且其相关值在搜索窗的起始位置， 这时过釆样 会导致前一个搜索窗内也出现峰值， 该峰值的出现就是一种虚警， 如图 2中 所示， 由于釆用了过釆样技术， 用户 2的信号能量被分散到前一个搜索窗了， 导致 D点也超过检测门限， 形成虚警。
RACH信号的检测捕获性能可以用前导的漏检率、 虚警率以及检测门限 来表征。 在给定虚警率目标的前提下， 通过确定对应的检测门限， 可以测试 RACH信号的漏检率。 这里， 所述虚警率定义为在时域相关检测中， 当没有 信号发送时检测到的前导的概率。 所述漏检率定义为当检测到一个不同于发 射的前导码， 或者根本没有检测到一个前导码， 或者检测到正确的前导但是 错误的定时估计等这些情况发生的概率。
一般情况下， 要求虚警率的目标为 10— 3或其以下但邻近 10- 3 , 规定在测试 中在没有信号发送时达到虚警率为 10— 3或其以下但邻近 10- 3时的检测门限为 RACH检测性能的门限。
一般来说， 经过 RACH峰值检测的绝对门限与噪声功率的大小有关系， 但是， 以随机接入信号的相关值与噪声功率的比值作为门限时， 这个门限就 和噪声功率无关， 因此定义所述检测门限为信号相关值与噪声功率的比值为 相对门限。 当接收信号的相关值与噪声功率的比值大于相对门限时， 认为有 RACH信号存在。 在正常情况下， 所述相对门限值设得越高， 用户的移动终 端就越难接入； 反之， 若该相对门限值设的越低， 则虚警就越多。 通常状况 下， 所述相对门限取虚警率为 10— 3时的值。这个相对门限值是在不发送信号时 测得的， 因此在 RACH前导检测时， 相对门限已经预先设置好了。
确定相对门限值后， 还需要估计噪声功率才能根据所收到的相关值对 RACH信号进行检测。所述噪声功率的估计是对没有 RACH信号的相关数据， 即噪声以及干扰数据做平均得到的， 因此还需要去掉那些 RACH信号的相关 值。 对于有同一个才艮序列的用于前导的多个序列循环移位被不同 RACH用户 使用的情况， 由于接收到的各个 RACH用户的功率不同， 实现准确快速估计 噪声功率是比较困难的。 噪声功率估计不准确将会影响检测门限的取值， 而 检测门限过低会导致虚警增加， 过高会导致漏检增加。 而且， 若对每一个时 域相关值都去判断是否为噪声数据， 处理起来复杂度较高且延时较大。
在实际的无线环境中， 接收到的信号是经过多条传播路径的信号， 每条 传播路径上的信号到达接收机的延时和功率也不同， 这叫做多径效应。 由于 多径效应， 在一个搜索窗内会出现同一个 RACH信号的多个不同延时的相关 值点， 当有多个多径信号相关值点超过门限时， 在超过门限的点中如何选择 一个合理的定时位置也不确定。
综上所述，现有 LTE系统对移动终端通过 RACH信道完成上行同步接入 系统的处理中， 存在不能快速准确地估计噪声功率所引起的接入成功率变低 或虚警率变高、 数据处理的复杂度较高等问题， 还存在由于釆用过釆样技术 所引起的虚警率高、 所得到的上行定时调整信息准确性不足的缺陷。
发明内容
有鉴于此， 本发明的主要目的在于提供一种随机接入前导的检测方法， 为移动终端提供准确可靠的上行定时调整信息， 以提高移动终端接入移动通 信系统的可靠性， 并降低处理虚警引起的信令开销。
为达到上述目的， 本发明的技术方案是这样实现的：
一种随机接入前导的检测方法， 该方法包括：
A、 对经过随机接入信道 RACH前导部分的时域相关处理后得到的所选 的本地根序列的时域 RACH信号进行噪声均值估计及绝对门限的计算；
B、从所选的本地根序列对应的可用搜索窗中选出一个没有被检测过的可 用搜索窗作为当前搜索窗， 对所述当前搜索窗内的时域相关值进行检测， 得 到检测结果值，将所述检测结果值及其对应的时域位置送入步骤 C进行处理； c、 根据预置的 选条件对所述检测结果值及其对应的时域位置进行判 断， 确定 RACH前导信号的定时位置。
进一步地， 步骤 A之前， 该方法进一步包括： 对接收到的前导信号进行 降釆样处理， 并进行 FFT处理， 提取其中的频域 RACH信号后， 从当前小区 分配的根序列中选出一条没有被检测过的本地根序列。
进一步地， 步骤 A中所述时域相关处理具体为： 将其与所选的本地根序 列复共轭点乘， 然后再经过 IFFT处理将所述信号变换到时域。
进一步地，步骤 A中所述对时域 RACH信号进行噪声均值估计及对绝对 门限的计算具体为： 查找每个 RACH信号根序列的所有时域相关值中的最大 值， 对小于最大值 nl倍的相关值求均值， 作为临时噪声均值， 再以所述临时 噪声均值乘以 n2倍的预设的相对门限得到噪声门限值，然后对小于所述噪声 门限的相关值求均值， 得到所述序列的噪声均值， 最后， 以所述噪声均值与 所述相对门限值相乘的积作为绝对门限值； 其中， nl和 n2都是小于 1的正 数。
较佳地， 步骤 B中所述得到检测结果值具体为： 查找每个搜索窗内的最 大相关值， 当最大相关值超过绝对门限值时， 对所述搜索窗内每个超过绝对 门限值的点及其左右相邻的若干个点对应的相关值进行平均， 当所得的平均 值仍然超过绝对门限时， 将所述超过绝对门限值的相关值作为检测结果值。
较佳地， 所述步骤 B中对所述当前搜索窗内的时域相关值进行检测时， 若得不到检测结果值， 则所述搜索窗内没有 RACH信号。
进一步地， 步骤 C中所述预置的筛选条件之一为： 所述检测结果值的 m 倍大于所述当前搜索窗的最大相关值， 其中： m为大于 1的正数；
进一步地， 步骤 C中所述预置的筛选条件之二为：
在所述当前搜索窗尾部 k个点中的最大值小于当前超过绝对门限的点与 其两边相邻的若干点相关值的均值或者超过绝对门限的点的位置不在搜索窗 尾部 k个点的范围内， 其中， k为正整数。 进一步地， 步骤 C中所述预置的筛选条件之三为：
以所述当前搜索窗内与第一个超过绝对门限的点的时域距离小于一定时 长、 且所述结果值中相关值最大的点所对应的位置作为定时位置。
较佳地， 步骤 C具体为： 将满足所述三个预置的筛选条件的检测结果值 对应的时域位置作为确定的唯一的定时位置； 否则， 判定该搜索窗内没有 RACH信号。
本发明所提供的随机接入前导的检测方法， 具有以下优点：
1 )本发明通过对移动终端所接收到的信号进行处理， 并分别以 FFT和 IFFT处理代替 DFT和 IDFT处理过程，降低了提取所述信号中频域 RACH信 号以及与本地频域根序列复共轭点乘计算的运算量及数据处理难度。
2 )本发明对经过 IFFT处理后的时域信号进行噪声均值估计和绝对门限 的计算， 降低了 RACH信号中的噪声干扰， 为系统进一步处理 RACH信号提 供了可靠的理论依据， 同时， 设置合理的绝对门限值， 降低了系统对前导码 的漏检率和虚警率。
3 )本发明通过寻找每个搜索窗内的最大相关值， 并对最大相关值及其左 右相邻的若干点进行平均计算， 使系统能够快速、 准确地寻找到有 RACH信 号的搜索窗， 并减少了虚警的发生。
4 )本发明对定时位置进行确定时， 通过设置定时位置的三个筛选条件， 有效抑制了由于使用过釆样技术或由于噪声引起的虚警， 从而降低了虚警所 引起的不必要的信令开销； 另外， 通过对可能是定时点的几个时域相关值的 平均， 可以比较准确地得到定时点， 消除过釆样引起的定时误差； 并且通过 对超过绝对门限的检测结果值时域位置的距离的限制， 能够保证定时位置位 于能量较大且定时偏差较小的位置。 附图概述
图 1为现有的 RACH时域相关检测的方法示意图；
图 2为 RACH峰值检测的数据处理过程示意图；
图 3为本发明实施例的 RACH时域相关检测过程流程图。 本发明的较佳实施方式
下面结合附图及本发明的实施例对本发明的方法作进一步详细的说明。 在 LTE标准中，规定 RACH信号在频域上占用的带宽为 1.08MHz,规定 每个 RACH子载波间隔为 1.25KHZ。这样，在该带宽范围内共有 864个 RACH 子载波， 其中 839个子载波放置 RACH数据， 其余 25个子载波作为保护带。 在 LTE标准中还规定 RACH信号的时域格式为：循环前缀（CP, Cycle Prefix ) 后面跟随 RACH的前导， RACH的前导后面是保护间隔。 其中， 保护间隔内 不发送任何信号。
下面才艮据已经测得的虚警率目标为 10— 3时的相对检测门限， 以 30.72MHz 釆样频率的 RACH信号为例来说明整个 RACH信号的检测过程。
图 3为本发明实施例 RACH时域相关检测过程流程图， 如图 3所示， 该 过程包括：
步骤 301 : 基站提取出接收到的信号的前导部分后进行降釆样， 并进行 FFT处理， 提取频域数据中的 RACH信号。
具体为： 对接收到的信号去 CP, 然后从 30.72MHz降釆样到 1.28MHz, 接着进行 1024点的 FFT处理， 提取出 839点的 RACH频域信号；
步骤 302: 判断分配给当前小区的所有根序列是否都已经检测过， 若全 部已检测过， 则执行步骤 311 ; 否则执行步骤 303。
步骤 303 : 进行根序列相关检测， 并从当前小区分配的根序列中选出一 条没有被检测过的根序列。
这里， 所述被选出的根序列， 即本地频域根序列。
步骤 304: 将所接收到的频域 RACH信号与所选的本地频域根序列复共 轭点乘， 然后再经过 IFFT处理将所述 RACH信号变换到时域， 得到时域相 关数据。
具体为： 将所提取的 839点的 RACH频域信号和所选的本地频域根序列 复共轭点乘， 补零到 2048点， 再进行 2048点的 IFFT处理， 得到所选的本地 根序列的 2048个时域相关点的数据； 然后对所述 2048点的每个点求模平方， 得到所述 2048点的时域相关数据， 所述 2048点时域相关数据实际上是数据 功率值。
步骤 305: 所述基站对通过 RACH时域相关处理后得到的所选的本地根 序列的时域 RACH信号进行噪声均值估计及绝对门限的计算。
这里， 所述噪声均值估计及绝对门限的计算， 具体为： 寻找每个 RACH 信号根序列的所有时域相关值中的最大值， 对小于最大值 n ( 0 < "≤1 )倍的 数据求均值， 将所求得的平均值作为临时噪声均值。 再将所述临时噪声均值 乘以 n ( 0 < "≤1 )倍的相对门限值， 得到噪声门限值。 通常将小于噪声门限 的相关值数据视为不包含 RACH信号的噪声干扰数据， 这些小于噪声门限的 相关值数据的均值即为该根序列的噪声功率均值， 噪声功率均值简称为噪声 均值。 所述噪声均值与相对门限值的乘积称为绝对门限。
例如： 在得到 2048点的时域相关数据后， 从中找出最大值， 即如图 2中 所示， 最大值点为 B点， 此时 n取 3/5 , 则通过计算小于最大值 3/5倍的相关 数据的平均值， 得到临时噪声均值； 再将临时噪声均值乘以 3/5倍的相对门 限，得到噪声门限值。此时，将小于噪声门限值的相关数据视为不包含 RACH 信号的噪声干扰数据， 计算这些小于噪声门限的相关值数据的均值即得到该 根序列的噪声均值。
可见， 通过该方法可以快速、 准确地计算出噪声功率均值即噪声均值。 将噪声均值与虚警率为 10-3时的相对检测门限的乘积作为绝对检测门限值，绝 对检测门限值简称绝对门限值。
步骤 306: 判断所选的本地根序列对应的所有可用搜索窗是否都已经检 测过， 如果都检测过， 则执行步骤 302, 进行下一条根序列的检测； 否则执 行步骤 307 , 即从所选的本地根序列对应的可用搜索窗中选出一个没有被检 测过的可用搜索窗进行检测。
步骤 307: 从所选的本地根序列对应的可用搜索窗中选出一个没有检测 过的可用搜索窗， 确定所选的没有检测过的可用搜索窗的起始和终止位置。
这里， 所选的没有检测过的可用搜索窗位置的确定， 具体为： 在 LTE标 准中给出了每条根序列使用的搜索窗长度 Ncs值， 以及可计算出每条根序列 使用的循环移位序列的循环移位公式， 并限定每个小区有 64条序列， 这 64 条序列可以由多个根序列和根序列的不同循环移位序列组成， 通过这些信息 可以确定每个本地根序列对应的每个可用搜索窗的位置。
例如， 某个低速小区使用的搜索窗长度 Ncs为 13 , 根序列使用的是第 2 条根序列， 则该小区的 64条前导序列由第 2条根序列的 64个循环移位序列 组成， 该根序列中有 64个搜索窗， 其中第一个搜索窗位置是从第 1个样点到 第 13个样点， 如果考虑 2048点 IFFT引起的过釆样， 则第一个搜索窗的位置 是从第 1个样点到第 32个样点。
步骤 308: 对所选的没有检测过的可用搜索窗内的信号相关值进行检测， 当所选的没有检测过的可用搜索窗内的最大相关值点超过绝对门限， 且任意 一个超过绝对门限的点及其左右各相邻的几个点的相关值的平均值仍然超过 绝对门限时， 执行步骤 309; 否则， 认为当前搜索窗内没有可用的信号， 返 回步骤 306。
其中， 疑似信号点是指搜索窗内每个超过绝对门限的点及其左右各相邻 的几个点。
这里， 对所选的没有检测过的可用搜索窗内的信号进行检测， 具体为： 寻找所选的没有检测过的可用搜索窗内的最大相关值点， 当最大相关值点超 过绝对门限时， 对该可用搜索窗内每个超过绝对门限的点及其左右各相邻的 几个点的相关值进行平均得到平均值， 当所得的任意一个超过绝对门限的点 及其左右各相邻的几个点的相关值的平均值仍然超过绝对门限时， 则执行步 骤 309; 否则， 认为所选的没有检测过的可用搜索窗内每个点均不满足执行 步骤 309的条件，即认为所选的没有检测过的可用搜索窗内没有可用的信号， 则退出本次的信号检测过程， 返回步骤 306, 开始对下一个搜索窗内的信号 进行检测。
例如： 对每个循环移位对应的搜索窗， 搜索其内的最大相关值， 当最大 相关值超过绝对门限时， 对该搜索窗内每个超过绝对门限的点以及其两边各 相邻的 1个点进行平均得到平均值， 当所得的平均值仍然超过所述绝对门限 值， 执行步骤 309 , 进行定时位置的确定过程。 当该搜索窗内的最大相关值 没有超过绝对门限； 或该搜索窗内最大相关值超过绝对门限， 但每个过绝对 门限的点以及其两边各相邻的 1个点的平均值没有超过绝对门限时， 说明没 有 RACH信号。
这里， 所述每个超过绝对门限的点的两边相邻的点， 是按照时间顺序罗 列在该点两侧且与该点相邻的相关值点， 其既可以是超过绝对门限的点， 也 可以是不超过绝对门限的点。
釆用上述处理策略可以保证快速找到当前有 RACH的搜索窗。 如图 2所 示的时域相关数据， 根据上面的处理方法， 可以很快判断出 A点、 B点、 C 点、 D点、 E点、 F点、 G点以及 H点所在的搜索窗内可能有的信号； 这样 就避免了由于釆用了过釆样技术而造成的非信号点被检测到而形成虚警， 并 保证了实际的真正的定时点的相关值不被遗漏。
步骤 309: 根据设定的筛选条件确定发送前导起点的定时位置， 判断所 选的没有检测过的可用搜索窗内的任一超过绝对门限的点是否符合筛选条 件， 若符合， 则执行步骤 310; 否则， 返回步骤 306, 表示所述每个点的值都 不能满足定时位置条件时， 也认为当前搜索窗内没有可用的信号， 即退出当 前搜索窗的信号检测过程， 对下一个没有检测过的搜索窗进行检测。 。
具体为： 当当前搜索窗内的任一超过绝对门限的点同时满足下面三个条 件时， 系统即认为找到了定时位置， 否则判定该点不是定时位置。 其中， 需 要同时满足的三个条件如下：
条件一是所选的没有检测过的可用搜索窗内的超过绝对门限的点的 m倍 大于该搜索窗的最大相关值， 其中： m为大于 1的正数； 条件二是在搜索窗 尾部 k个点中的最大值小于当前超过绝对门限的点与其两边相邻的若干点相 关值的均值， 或超过绝对门限的点的位置不在搜索窗尾部 k个点的范围内， 其中， k为正整数； 条件三是以当前搜索窗内与第一个超过绝对门限的点的 时域距离小于一定时长、 且超过绝对门限的点中相关值最大的点所对应的位 置为定时位置。
例如： 对于进入定时位置的确定的每个点， 当同时满足如下条件时， 就 认为找到定时点位置， 否则认为该点不是定时点。 条件一： 当超过绝对门限 的点对应的相关值的 3倍的值大于该搜索窗的最大相关值； 条件二： 在搜索 窗尾部 5个点中的最大值小于该超过绝对门限的点与其左右相邻一点相关值 的均值或该超过绝对门限的点的位置不在搜索窗尾部 5个点范围内；条件三： 以当前搜索窗内与第一个超过绝对门限的相关值点的时域距离小于一定时 长， 且超过绝对门限的检测结果值中相关值最大的点所对应的位置作为定时 位置。 这里， 所述一定时长， 可设为 2.56微秒。
因为， 满足条件一可以消除噪声干扰引起的虚警以及在高信噪比时由于 过釆样造成的定时误差，如图 2所示， 对用户 1而言， A点、 B点、 C点都有 可能是定时点， 但满足定时条件一的只有 B点， 实际上也只有 B点才是用户 1 的真实定时点。 满足条件二则可以消除由于过釆样引起的本搜索窗的相关 峰值漂移到邻近的搜索窗而造成虚警， 如图 2所示， 由于用户 2的信号能量 分散到相邻的一个搜索窗即前一个搜索窗了， 导致 D点超过绝对门限， 但该 点位于当前搜索窗的尾部的 5个点内， 于是可以肯定地认为该相邻的搜索窗 内没有 RACH信号， 从而有效抑制了处理虚警给系统带来的使得信令开销增 大的不利影响。 满足条件三可以保证定时位置位于能量较大且定时偏差较小 的位置。 如在有信号的 RACH搜索窗内， 因为多径的影响， 有多个过门限且 满足筛选条件一和二的检测点， 最后确定以与第一个过门限的检测点的时域 距离小于 2.56微秒且超过绝对门限的检测结果值中相关值最大的检测点对应 的位置为定时位置。
步骤 310: 当所选的没有检测过的可用搜索窗内的所有超过绝对门限的 点中有满足三个筛选条件的点时， 把满足三个筛选条件的点对应的定时位置 转换成定时调整量， 为发送本前导的用户提供上行同步时间提前信息。
步骤 311 : 当本小区分配的所有前导序列都检测过后， 结束本小区的 RACH前导检测过程。
通过使用本发明提供的 RACH信号的检测方法， 可以对每条根序列均进 行一个有效的噪声估计， 并根据估计的噪声得到的信号检测绝对门限， 对基 于同一根序列的不同循环移位的 RACH用户均适用，由于信号检测门限准确、 实时， 因此可以快速、 准确地检测到 RACH用户； 本发明提供的检测方法还 考虑了对因过釆样或噪声引起的虚警的抑制， 减少了虚警带来的不必要的信 令开销； 通过对可能是定时点的几个时域相关值的平均， 可以比较准确地得 到定时点并减少虚警的发生， 消除了过釆样引起的定时误差， 对超过绝对门 限的检测结果值时域位置的距离的限制， 可以保证定时位置位于能量较大且 定时偏差较小的位置， 使用户到达基站时的信号功率值较高且定时误差小， 保证了后续上行数据的可靠传输。
以上所述， 为本发明的较佳实施例而已， 并非用于限定本发明的保护范 围。
工业实用性
本发明通过对移动终端所接收到的信号进行处理，并分别以 FFT和 IFFT 处理代替 DFT和 IDFT处理过程， 降低了提取所述信号中频域 RACH信号以 及与本地频域根序列复共轭点乘计算的运算量及数据处理难度； 对经过 IFFT 处理后的时域信号进行噪声均值估计和绝对门限的计算， 降低了 RACH信号 中的噪声干扰，为系统进一步处理 RACH信号提供了可靠的理论依据， 同时， 设置合理的绝对门限值， 降低了系统对前导码的漏检率和虚警率； 通过寻找 每个搜索窗内的最大相关值， 并对最大相关值及其左右相邻的若干点进行平 均计算， 使系统能够快速、 准确地寻找到有 RACH信号的搜索窗， 并减少了 虚警的发生； 对定时位置进行确定时， 通过设置定时位置的三个筛选条件， 有效抑制了由于使用过釆样技术或由于噪声引起的虚警， 从而降低了虚警所 引起的不必要的信令开销； 另外， 通过对可能是定时点的几个时域相关值的 平均， 可以比较准确地得到定时点， 消除过釆样引起的定时误差； 并且通过 对超过绝对门限的检测结果值时域位置的距离的限制， 能够保证定时位置位 于能量较大且定时偏差较小的位置。 由此可见， 本发明具有极强的工业实用 性。

Claims

权 利 要 求 书
1、 一种随机接入前导的检测方法， 该方法包括：
A、 对经过随机接入信道 RACH前导部分的时域相关处理后得到的所选 的本地根序列的时域 RACH信号进行噪声均值估计及绝对门限的计算；
B、从所选的本地根序列对应的可用搜索窗中选出一个没有被检测过的可 用搜索窗作为当前搜索窗， 对所述当前搜索窗内的时域相关值进行检测， 得 到检测结果值，将所述检测结果值及其对应的时域位置送入步骤 C进行处理；
C、 根据预置的 选条件对所述检测结果值及其对应的时域位置进行判 断， 确定 RACH前导信号的定时位置。
2、 根据权利要求 1所述的方法， 其中， 步骤 A之前， 该方法进一步包 括： 对接收到的前导信号进行降釆样处理， 并进行 FFT处理， 提取其中的频 域 RACH信号后， 从当前小区分配的根序列中选出一条没有被检测过的本地 根序列。
3、 根据权利要求 1所述的方法， 其中， 步骤 A中所述时域相关处理具 体为： 将其与所选的本地根序列复共轭点乘， 然后再经过 IFFT处理将所述信 号变换到时域。
4、 根据权利要求 1所述的方法， 其中， 步骤 A中所述对时域 RACH信 号进行噪声均值估计及对绝对门限的计算具体为： 查找每个 RACH信号根序 列的所有时域相关值中的最大值， 对小于最大值 nl倍的相关值求均值， 作为 临时噪声均值，再以所述临时噪声均值乘以 n2倍的预设的相对门限得到噪声 门限值， 然后对小于所述噪声门限的相关值求均值， 得到所述序列的噪声均 中， nl和 n2都是小于 1的正数。
5、根据权利要求 1所述的方法， 其中， 步骤 B中所述得到检测结果值具 体为： 查找每个搜索窗内的最大相关值， 当最大相关值超过绝对门限值时， 对所述搜索窗内每个超过绝对门限值的点及其左右相邻的若干个点对应的相 关值进行平均， 当所得的平均值仍然超过绝对门限时， 将所述超过绝对门限 值的相关值作为检测结果值。
6、根据权利要求 1所述的方法， 其中， 所述步骤 B中对所述当前搜索窗 内的时域相关值进行检测时， 若得不到检测结果值， 则所述搜索窗内没有 RACH信号。
7、根据权利要求 5所述的方法， 其中， 步骤 C中所述预置的筛选条件之 一为： 所述检测结果值的 m倍大于所述当前搜索窗的最大相关值， 其中： m 为大于 1的正数；
8、根据权利要求 5所述的方法， 其中， 步骤 C中所述预置的筛选条件之 二为：
在所述当前搜索窗尾部 k个点中的最大值小于当前超过绝对门限的点与 其两边相邻的若干点相关值的均值或者超过绝对门限的点的位置不在搜索窗 尾部 k个点的范围内， 其中， k为正整数。
9、根据权利要求 5所述的方法， 其中， 步骤 C中所述预置的筛选条件之 三为：
以所述当前搜索窗内与第一个超过绝对门限的点的时域距离小于一定时 长、 且所述结果值中相关值最大的点所对应的位置作为定时位置。
10、 根据权利要求 7 ~ 9中任一项所述的方法， 其中， 步骤 C具体为： 将 满足所述三个预置的筛选条件的检测结果值对应的时域位置作为确定的唯一 的定时位置； 否则， 判定该搜索窗内没有 RACH信号。
PCT/CN2009/072486 2008-06-27 2009-06-26 一种随机接入前导的检测方法 WO2009155882A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN200810126188.9 2008-06-27
CN200810126188.9A CN101295999B (zh) 2008-06-27 2008-06-27 一种随机接入前导的检测方法

Publications (1)

Publication Number Publication Date
WO2009155882A1 true WO2009155882A1 (zh) 2009-12-30

Family

ID=40066055

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2009/072486 WO2009155882A1 (zh) 2008-06-27 2009-06-26 一种随机接入前导的检测方法

Country Status (2)

Country Link
CN (1) CN101295999B (zh)
WO (1) WO2009155882A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108683482A (zh) * 2017-04-01 2018-10-19 电信科学技术研究院 一种估计定时位置的方法及装置
CN109871509A (zh) * 2019-02-19 2019-06-11 哈尔滨工程大学 一种基于ar算法的瞬态信号高分辨检测方法

Families Citing this family (35)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101295999B (zh) * 2008-06-27 2014-02-19 中兴通讯股份有限公司 一种随机接入前导的检测方法
CN101534542B (zh) * 2009-04-16 2011-11-02 华为技术有限公司 随机接入前导序列的调整方法、装置及系统
CN101944929B (zh) * 2009-07-06 2013-08-07 中兴通讯股份有限公司 基于随机接入过程的检测方法和检测装置
CN101997590B (zh) * 2009-08-10 2014-06-11 中兴通讯股份有限公司 随机接入信号的检测方法和基带处理板以及基站
CN102065535B (zh) * 2009-11-11 2014-07-23 电信科学技术研究院 一种随机接入过程中定时提前量的确定方法及装置
CN102075960A (zh) * 2009-11-25 2011-05-25 中兴通讯股份有限公司 一种随机接入信号的检测方法和基站
CN102111894B (zh) * 2009-12-24 2014-08-13 中兴通讯股份有限公司 随机接入信道中抑制频偏虚警的方法及装置
CN101820301B (zh) * 2010-04-27 2012-12-19 安徽创毅通信科技有限公司 一种长期演进系统中随机接入前导低复杂度生成方法
CN102612051B (zh) * 2011-01-24 2014-11-05 中兴通讯股份有限公司 基于td-lte的随机接入检测方法及装置
CN102892207B (zh) * 2011-07-22 2015-08-19 普天信息技术研究院有限公司 一种物理随机接入信道的检测方法
CN103037392B (zh) * 2011-09-30 2015-12-02 京信通信系统（中国）有限公司 Lte前导信号的检测方法、装置及基站
CN102404724B (zh) * 2011-10-28 2015-07-08 华为技术有限公司 用户筛选方法及用于用户筛选的基站
CN103108338B (zh) * 2011-11-14 2015-06-17 京信通信系统（中国）有限公司 一种随机接入信号的检测方法、装置及基站
CN102547831A (zh) * 2012-01-09 2012-07-04 大唐移动通信设备有限公司 一种随机接入检测方法及装置
CN102447532B (zh) * 2012-01-19 2014-05-28 大唐移动通信设备有限公司 一种前导码检测方法及装置
US8837312B2 (en) 2012-02-29 2014-09-16 Hong Kong Applied Science And Technology Research Institute Co., Ltd. Signature detection and timing offset estimation
US9001678B2 (en) * 2012-02-29 2015-04-07 Hong Kong Applied Science And Technology Research Institute Co., Ltd. False alarm reduction with search windowing and peak suppression
CN103313274B (zh) * 2012-03-13 2016-12-14 普天信息技术研究院有限公司 一种抑制随机接入码虚警的方法
CN103687046B (zh) * 2012-09-26 2017-03-22 京信通信系统（中国）有限公司 一种随机接入前导序列的检测方法及装置
CN102957658B (zh) * 2012-11-08 2015-02-18 西华大学 一种ofdm通信系统的时间同步方法及装置
CN103079227B (zh) * 2013-02-05 2015-04-29 武汉邮电科学研究院 一种用于lte系统的随机接入检测方法和系统
WO2015001369A1 (en) * 2013-07-01 2015-01-08 Freescale Semiconductor, Inc. Radio signal decoding and decoder
CN104918336A (zh) * 2014-03-12 2015-09-16 中兴通讯股份有限公司 一种干扰环境下随机接入信号的检测方法、装置和系统
CN103929825A (zh) * 2014-04-30 2014-07-16 电子科技大学 基于zc序列的多用户检测方法
CN105099965B (zh) * 2014-05-12 2018-01-02 普天信息技术有限公司 一种随机接入信道的噪声方差估计方法及装置
CN104076187B (zh) * 2014-06-10 2016-12-07 深圳市豪恩声学股份有限公司 压电信号峰值检测方法及其装置
CN105306097B (zh) * 2014-06-18 2019-07-26 中兴通讯股份有限公司 一种随机接入信号的检测方法、装置和系统
US9913295B2 (en) * 2016-06-24 2018-03-06 Alcatel-Lucent Usa Inc. System and method of controlling network traffic using preamble detection and automated optimization of random access channel parameters
CN108243140A (zh) * 2017-08-14 2018-07-03 张涛 一种前导信号检测方法和装置
WO2019127396A1 (en) * 2017-12-29 2019-07-04 Nokia Shanghai Bell Co., Ltd. Method and device for nprach detection
WO2019232726A1 (en) * 2018-06-06 2019-12-12 Nokia Shanghai Bell Co., Ltd. Methods, device and computer-readable medium for determining timing advance
CN111107562B (zh) * 2018-10-29 2021-05-28 大唐移动通信设备有限公司 一种检测方法和检测装置
CN109474920A (zh) * 2018-11-02 2019-03-15 上海华虹集成电路有限责任公司 一种低功耗蓝牙数据包的检测方法
CN114731701B (zh) * 2019-12-04 2023-11-28 华为技术有限公司 一种序列检测方法及设备
CN113038514B (zh) * 2021-03-24 2022-09-13 中国科学院计算技术研究所 一种5g prach峰值检测算法的自适应方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1330813A (zh) * 1998-12-14 2002-01-09 交互数字技术公司 随机接入信道前同步信号的检测
CN101098218A (zh) * 2006-06-27 2008-01-02 中兴通讯股份有限公司 一种随机接入前导信号的发射机和接收机
WO2008025373A1 (en) * 2006-08-29 2008-03-06 Telefonaktiebolaget Lm Ericsson (Publ) Detection of access bursts in a random access channel
US20080095254A1 (en) * 2006-10-24 2008-04-24 Tarik Muharemovic Random Access Channel Design With Hybrid CDM And FDM Multiplexing Of Access
CN101295999A (zh) * 2008-06-27 2008-10-29 中兴通讯股份有限公司 一种随机接入前导的检测方法

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN100444684C (zh) * 2004-08-04 2008-12-17 中兴通讯股份有限公司 码分多址系统随机接入前导检测动态门限方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1330813A (zh) * 1998-12-14 2002-01-09 交互数字技术公司 随机接入信道前同步信号的检测
CN101098218A (zh) * 2006-06-27 2008-01-02 中兴通讯股份有限公司 一种随机接入前导信号的发射机和接收机
WO2008025373A1 (en) * 2006-08-29 2008-03-06 Telefonaktiebolaget Lm Ericsson (Publ) Detection of access bursts in a random access channel
US20080095254A1 (en) * 2006-10-24 2008-04-24 Tarik Muharemovic Random Access Channel Design With Hybrid CDM And FDM Multiplexing Of Access
CN101295999A (zh) * 2008-06-27 2008-10-29 中兴通讯股份有限公司 一种随机接入前导的检测方法

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108683482A (zh) * 2017-04-01 2018-10-19 电信科学技术研究院 一种估计定时位置的方法及装置
EP3609110A4 (en) * 2017-04-01 2020-04-08 China Academy of Telecommunications Technology METHOD AND DEVICE FOR ESTIMATING THE LOCATION OF SYNCHRONIZATION
CN108683482B (zh) * 2017-04-01 2021-03-09 电信科学技术研究院 一种估计定时位置的方法及装置
US11044134B2 (en) 2017-04-01 2021-06-22 China Academy Of Telecommunications Technology Method and device of estimating timing location of round trip time
CN109871509A (zh) * 2019-02-19 2019-06-11 哈尔滨工程大学 一种基于ar算法的瞬态信号高分辨检测方法

Also Published As

Publication number Publication date
CN101295999B (zh) 2014-02-19
CN101295999A (zh) 2008-10-29

Similar Documents

Publication Publication Date Title
WO2009155882A1 (zh) 一种随机接入前导的检测方法
EP2439973B1 (en) Detection method and apparatus based on random access process
US8165106B2 (en) Apparatus and method for detecting a ranging signal in a wireless communication system
US8155251B2 (en) Detection of access bursts in a random access channel
US9615386B2 (en) Random access channel false alarm control
US8218427B2 (en) Preamble configuring method in the wireless LAN system, and a method for a frame synchronization
US8391131B2 (en) Method for processing the random access transmission in the frequency domain
US9398617B2 (en) Methods and apparatus for random access in a communication system
WO2010028537A1 (zh) 一种随机接入信道的信号检测方法
CN108924943B (zh) 基于窄带物联网随机接入信道的最大相关估计检测方法
CN102209052B (zh) 一种随机接入信号的频偏补偿方法和装置
KR101069988B1 (ko) 무선랜에서 신속한 동기 검출을 위한 상관 방법 및 상관 장치
US20150023329A1 (en) Wireless network signal to interference plus noise ratio estimation for a random access channel
JP2005303691A (ja) 同期検出装置および同期検出方法
EP3308567B1 (en) Systems and methods for detecting nodes or signals of a radio access technology (rat) by a node of another rat
WO2008022598A1 (fr) Procédé de test, terminal et dispositif côté réseau pour accès aléatoire
CN101820407B (zh) 基于串行干扰抵消的频域初始测距方法及系统
Figueiredo et al. LTE random access detection based on a CA-CFAR strategy
KR101257015B1 (ko) 타이밍 동기화를 위한 다중 단계 타이밍 옵셋 추정 방법 및 ｏｆｄｍ 수신기
US10212679B1 (en) Method and apparatus for delay spread estimation
KR100625408B1 (ko) Ｏｆｄｍａ／ｔｄｄ 통신 시스템의 하향링크 프리앰블 검출장치
KR101257014B1 (ko) 샘플 표준 편차를 기초로 한 다중 경로 페이딩 채널에서 ｏｆｄｍ 시스템을 위한 타이밍 옵셋 추정 방법 및 이러한 방법을 사용하는 장치
CN117222041A (zh) 5G小基站系统中PRACH信号preamble相关峰的检测方法和装置
KR20090032362A (ko) 이동통신시스템에서 자동 주파수 제어 장치 및 방법

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 09768807

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 09768807

Country of ref document: EP

Kind code of ref document: A1