CN103889069B - A kind of accidental access signal method of sending and receiving applied to large coverage - Google Patents

Abstract

The present invention provides a method for sending and receiving random access signals applied to a large coverage area. The steps are: the user equipment configures the RACH format, the RACH format is set by the CP length to include M repetitions of 24576Ts, and the T _SEQ length is set to N repetitions of 24576Ts, M=1, N≥1; R is the coverage scene radius, C is the speed of light, Ts=1/(15000*2048)s; the guard interval GT is adjusted by the RACH occupation length to be greater than or equal to the round-trip path delay; after sending the configured RACH signal, the base station extracts N +2 24576Ts long peak detection sequences; perform peak detection on N+2 peak detection sequences respectively; obtain the initial time offset value from the joint judgment of the leading ID results of each parallel detection and SINR; judge the same leading ID in In the case of N+1 parallel detection windows, the initial time offset value is calibrated to obtain the correct leading ID and time offset value; the invention can achieve a large-scale coverage of more than 100KM, and can meet the timing accuracy under ultra-long coverage Requirements, time-frequency resources are saved, and the modification of the protocol is minor.

Description

A random access signal transmission and reception method applied to a large coverage area

technical field

The invention relates to the field of frequency division duplex long-term evolution (FDD-LTE) of mobile communication systems, in particular to a method for sending and receiving random access signals under large coverage, and the invention belongs to the field of communication.

Background technique

In the LTE-FDD system, the user equipment establishes uplink initial synchronization with the base station through the random access process, and the initial timing advance (TA) of the user equipment is reported through random access channel measurement. Whether the uplink transmission time synchronization can be obtained, and whether it can be correctly scheduled for uplink transmission later has an important impact.

At present, there are four types of preamble formats for Random Access Channel (RACH) in LTE-FDD system, each format has its own cyclic prefix (CP) and sequence (T _SEQ ) length, and different CP and T _SEQ lengths have their own The corresponding reachable cell radius and delay range. Among the preamble formats that have been designed, the CP length of the preamble format 3 is 21024Ts, and the T _SEQ is composed of two 24576Ts long repetitive sequences. Using the traditional preamble detection method, the maximum delay that this format can support is 24576Ts, corresponding to about 100KM The coverage radius is not suitable for special scenarios with a radius of more than 100KM without adding base station equipment. In special scenarios with a radius of more than 100KM, using the existing preamble format and its traditional detection algorithm will lead to TA estimation errors. It will affect the success rate of the user equipment to obtain uplink synchronization.

In order to cover a larger area, the traditional transmission method needs to configure the corresponding CP and guard interval (GT) length of RACH according to the coverage requirements to offset the round-trip delay, that is, the larger the cell coverage, the longer the transmission delay, and the required CP and GT The larger the T _SEQ length and the frequency domain bandwidth of the RACH need to be increased at the same time, so that the RACH can meet the timing accuracy requirements of the system, resulting in a waste of time-frequency resources.

Contents of the invention

The purpose of the present invention is to provide a random signal access sending and receiving method with improved coverage, so as to save system time-frequency resources while expanding the coverage.

The technical scheme adopted in the present invention is:

A method for sending and receiving random access signals applied to a large coverage area, comprising the following steps: user equipment configures a random access channel RACH format, the CP length setting of the RACH format includes M repeated 24576Ts, and the T _SEQ length is set to N Repeated 24576Ts, where M=1, N≥1; N is determined by the formula R is the radius of the coverage scene, C is the speed of light, obtained by Ts= ^{1/(15000*2048)} _s ; the guard interval GT of the RACH signal is adjusted by the RACH signal occupation length to be greater than or equal to the round-trip path delay; the user sets and sends the configured RACH The signal is sent to the base station, and the base station extracts N+2 24576Ts long peak detection sequences; performs peak detection on the N+2 peak detection sequences respectively; obtains the initial time offset value for the joint judgment of the leading ID result of each parallel detection and the SINR ; By judging the appearance of the same preamble ID in N+1 parallel detection windows, the initial time offset value is calibrated to obtain the correct preamble ID and time offset value, and the random access signal is received successfully.

The extraction process of the peak detection sequence is as follows: the extracted sequence is divided into the first segment, the second segment...the segment sequence of the N+2 segment with a length of 24576Ts from the beginning to the end, and the parallel peak value is performed on the segment sequence respectively. Detection, record the leading sequence ID and time offset results of each peak detection and the SINR in the detection window of the N+1 and N+2 segments.

The process of obtaining the initial time offset value is as follows: remove the detection result of the N+2th segment in the segment sequence, and when the same leading ID is detected in the N+1th segment and at least one of the remaining N segments, then the N+1th segment The time offset value estimated by the segment sequence is used as the initial time offset value; the detection result of the N+2 segment in the segment sequence is removed, and when the same leading ID only appears in the N+1 segment, compare the N+1 segment with the N+ For the SINR of 2 segments, the time offset value estimated from the sequence with the larger SINR is taken as the initial time offset value.

The peak detection method of the N+2 sections of peak detection sequences is as follows: each section of the peak detection sequence is frequency-shifted to DC, the sequence is filtered and down-sampled to obtain a sequence with a length of 2048Ts, and a 2048-point fast Fourier transform is performed Transform, extract 839 points of the original ZC sequence from it, multiply the frequency domain sequence of the ZC sequence local mother code with the complex conjugate of the received RACH frequency domain sequence, and perform a discrete Fourier inverse transform of 1536 points to obtain the PDP time domain Correlation value: take the average value of the PDP time delay correlation value, set the threshold value and detect the peak value according to the PDP average value, record the detected preamble sequence ID and the corresponding time offset value of each segment, and calculate the SINR.

The time offset calibration includes three methods: 1) remove the detection result of the N+2 section, when the same preamble ID appears in the first N+1 parallel peak detections, the initial time offset value does not need to be calibrated, and the The initial time offset value is used as the last reported time offset value; II) The detection result of the N+2 segment is removed, when the same preamble ID is not detected in the 1...M (M=1...N) peak detection appears, and when it appears in the remaining N+1-M peak detections, add the initial time offset value to M*24576Ts to obtain the calibration value, and use the calibration value as the last reported time offset value; III) remove the N+th For the detection results of stage 2, when the same preamble ID does not appear in the N+1 peak detections, the time delay is too large and random access fails or the preamble ID is not sent. The SINR value is a ratio of the PDP peak energy in the detection window to the total PDP energy. The PDP peak energy is the PDP peak maximum energy plus the energy of points near the peak.

The present invention has the following advantages:

In order to successfully access in a large coverage scenario above 100KM, the present invention provides a random access method suitable for large coverage above 100KM. This method does not change the frequency domain resources occupied by RACH, and the additional time domain resources required are relatively Small, while ensuring system performance, saving system resources, and reducing overhead.

Description of drawings

Fig. 1 is the RACH preamble structure schematic diagram that the present invention designs;

Fig. 2 is a schematic diagram of extracting the RACH peak detection sequence of the present invention;

Fig. 3 is the implementation flowchart of the RACH transmission method of the present invention;

Fig. 4 is the RACH detection flowchart of the base station side of the present invention;

FIG. 5 is a flow chart of the RACH detection part at the base station side of the present invention;

FIG. 6 is a schematic structural diagram of the RACH preamble in Embodiment 1 of the present invention;

7 is a schematic diagram of extraction of RACH peak detection sequence in Embodiment 1 of the present invention;

FIG. 8 is a schematic structural diagram of the RACH preamble in Embodiment 2 of the present invention;

FIG. 9 is a schematic diagram of extraction of RACH peak detection sequence in Embodiment 2 of the present invention;

detailed description

The main realization principles, specific implementation methods and beneficial effects that can be achieved of the present invention will be described in detail below in combination with various embodiments.

The present invention relates to a random access signal sending and receiving method applicable to a large coverage area, including the following steps: the user equipment configures the random access channel RACH format, as shown in Figure 1, the CP length setting of the RACH format includes M Repeated 24576Ts, T _SEQ length is set to N repeated 24576Ts, where M=1, N≥1; N is determined by the formula R is the coverage scene radius, C is the speed of light, obtained by Ts= ¹ /( ^15000*2048 ) _s ; the guard interval GT of the RACH signal is adjusted by the RACH signal occupation length to be greater than or equal to the round-trip path delay; the user sets and sends the configured RACH The signal is sent to the base station, and the base station extracts N+2 24576Ts long peak detection sequences; performs peak detection on the N+2 peak detection sequences respectively; obtains the initial time offset value for the joint judgment of the leading ID result of each parallel detection and the SINR ; By judging the appearance of the same preamble ID in N+1 parallel detection windows, the initial time offset value is calibrated to obtain the correct preamble ID and time offset value, and the random access signal is received successfully.

The embodiments of the present invention are respectively described in detail for the coverage above 100KM and within 100KM. Embodiment 1 and Embodiment 2 are examples of coverage above 100KM, respectively, for coverage of 200KM and 360KM. Embodiment 3 is an example of coverage of 100KM.

Embodiment 1 of the present invention:

Under the condition of not changing the LTERACH frequency domain resource, the sending method and the receiving method of the present invention are described by taking the coverage of 200KM cells as an example.

In this embodiment, there are 3 users. In the low-speed scenario, the sent random access preamble ID numbers are 0, 1, 2, and Ncs=0. The path delay of each user in the embodiment is 1000Ts (within 100KM). 35000Ts (within 100～200KM), 50000Ts (over 200KM).

The sending method of embodiment 1 of the present invention will be described in detail below in conjunction with accompanying drawings 2 and 3, and the sending method includes the following steps:

Step 201, the user equipment configures RACH according to the following method;

In the case of not changing the frequency domain resources, at this time, a cell of 200KM is to be covered. The transmission method of the present invention is: first calculate the size of N, at this time R=200KM, according to the calculation formula of N Get N=1, T _SEQ length is 24576Ts, CP length is 24576Ts, round-trip path delay reaches 1.34ms, guard interval GT length is 1.34ms, then the total length under RACH is 2*24576Ts+1.34ms=1.6ms+1.34ms =2.94ms, only 3 TTIs are needed in total, since 3 TTIs are occupied in total, the GT length is adjusted to 1.4ms. The GT is designed according to the traditional method, that is, after the round-trip delay is calculated, the GT can be appropriately adjusted according to the length occupied by the RACH signal, and the guard interval GT is greater than or equal to the maximum round-trip delay.

The schematic diagram of the RACH preamble structure covered by 200KM in the embodiment of the present invention is shown in Figure 6. Using the sending and receiving method of the present invention, the required time-frequency resource overhead is the same as that of LTE's existing preamble format 3, which is specified in the protocol A RACH format with the largest coverage, its coverage is 100KM, and the sending and receiving method of the present invention can cover 200KM, effectively saves time-frequency resources, and expands the coverage of the system; at the same time, this preamble format is only in The RACH signal is changed in the time domain, while the frequency domain remains unchanged, and the protocol is changed slightly, which is convenient and easy to implement.

In particular, if the preamble format in this embodiment only needs to cover a range of 100KM, the receiving method of the present invention may not be used, and the traditional receiving method may be used, and the time-frequency resource overhead required by the existing preamble format 3 and the coverage Same range.

In step 202, a new RACH preamble is obtained, and an RACH signal is sent according to the RACH configuration mode.

The receiving method of embodiment 1 of the present invention to improve the coverage capability will be described in detail below in conjunction with accompanying drawing 4. The method for receiving random access signals includes the following steps:

Step 401, peak detection sequence extraction: the base station side receives the random access signal, obtains the time domain position of the preamble according to the preamble configuration index, and extracts (N+2)*24576Ts long data including CP, sequence and part of GT , since N=1 in this embodiment, then extract 3 long sequences of 24576Ts, and divide the data into 3 sections from front to back, as shown in Figure 7;

Step 402, performing parallel peak detection: performing parallel peak detection on the three peak detection sequences respectively. Record the preamble ID, time offset, and power delay profile (PDP) signal-to-noise ratio (SINR) for each detected segment.

This embodiment describes step 402 in detail in conjunction with FIG. 5:

Step 501, when performing peak detection on each sequence, shift the frequency of each peak detection sequence to direct current (DC), filter the sequence and down-sample by 12 times, obtain a sequence with a length of 2048Ts, and perform 2048-point fast Fu Liye transform (FFT), from which 839 points of the original Zadoff-Chu (ZC) sequence are extracted, and the frequency domain sequence of the local mother code of the ZC sequence is multiplied by the complex conjugate of the received RACH frequency domain sequence, and 1536 points are discrete Inverse Fourier transform (IDFT) to obtain the time-domain correlation value of the PDP.

Step 502, average the PDP, set the threshold value and detect the peak value according to the PDP average value, traverse 64 root sequences, detect the preamble ID and time offset of the user, and calculate and obtain the N+1 and SINR value of segment N+2. The method of estimating time bias based on the autocorrelation method is relatively mature, and will not be elaborated here. In this step, the calculated SINR for peak detection is the ratio of the PDP peak energy in the detection window to the total PDP energy. The PDP peak energy is the PDP peak maximum energy plus the energy of several points near the peak. In the present invention, the energy of 6 points near the peak can be selected, because the 839-point ZC sequence passes K points (for example, K=1536) IDFT to calculate PDP When the impact broadens, the energy is scattered.

Step 403, obtain initial time offset: obtain the initial time offset value through the joint judgment of the preamble ID result and SINR of each segment of parallel detection. Regardless of the detection result of the third paragraph: if the same preamble ID is not only detected in the second paragraph, that is, it is detected in both the first and second paragraphs, then the time offset value estimated in the second paragraph is used as the initial time offset value; if the same preamble ID only appears in the second segment, compare the SINR of the second segment and the third segment, and take the estimated time offset value of the sequence with a larger SINR as the initial time offset value; if the same preamble ID appears in the second segment Neither paragraph 1 nor paragraph 2 appears, and it is considered that there is no such user.

In this embodiment, the delays of three user paths are 1000Ts (within 100KM), 35000Ts (within 100-200KM), and 50000Ts (over 200KM). As shown in Figure 6, user 1 (with a delay of 1000Ts) falls within [0-24576) Ts and is detected in both the first and second sections, and the estimated time offset value in the second section is used as the initial time offset value. The time offset value is 1000Ts; user 2 (with a delay of 35000Ts) falls within [24576Ts～2*24576) Ts, and is only detected in the second segment. Comparing the SINR of the second segment and the third segment, the SINR of the third segment is larger , take the time offset value estimated in the third section as the initial time offset value, and the time offset value is 10424Ts; user 3 (with a delay of 60000Ts) exceeds 2*24576Ts, which does not appear in the peak detection of the first and second sections The preamble ID2 and the time offset of user 3 are too large and exceed the range of the preamble format time offset, and the random access of user 3 fails.

Step 404, perform calibration: by judging the occurrence of the leading ID of the same user in N+1 parallel detection windows, different time offset calibrations are performed respectively to obtain the correct time offset. The leading ID here is the leading ID of a user, and the leading IDs of different users are different. This step is the case where the same user is detected in the N+1 segment.

In this embodiment, regardless of the detection results in the third section, when the leading ID of user 1 appears in the two parallel peak detections in the first section and the second section, the initial time offset value does not need to be calibrated, and the initial time offset value as the last reported time offset value, that is, 1000TS as the last time offset value; when the leading ID of user 2 does not appear in the first peak detection but appears in the second peak detection, that is, it does not appear in the first segment but When the second paragraph appears, add 24576Ts to the initial time offset value to obtain a calibration value of 35000Ts, and use the calibration value as the last reported time offset value; when the leading ID of user 3 does not appear in the three peak detections, the delay exceeds A large value causes random access to fail. Therefore, in the case of N=1, the maximum time offset range that can be detected by this system is (1+1)*24576Ts=49152Ts.

For example, to cover a 200KM cell without changing the frequency domain resources, the round-trip path delay reaches 1.34ms at this time. Using the traditional design method and detection algorithm, regardless of the maximum delay extension, the CP and GT parts need at least 2.68ms. At the same time, T _SEQ needs to lengthen the 839RACH sequence change sampling to at least 1.34ms, which takes 4.02ms in total and takes up 5 TTIs. However, with the transmission method and detection method of the present invention, if N=1, the time delay of 2*24576Ts=1.6ms can be detected, the GT length is 1.34ms, the total length is 1.6+1.34=2.94ms, and 3 TTIs are enough to cover 200KM .

Embodiment 2 of the present invention:

Under the condition of not changing the LTERACH frequency domain resource, the sending method and the receiving method of the present invention are described by taking the coverage of 360KM cells as an example. In this embodiment, there are 4 users in total. In the low-speed scenario, the sent random access preamble ID numbers are 0, 1, 2, 3, Ncs=0, and the path delay added in the simulation is 1000Ts (within 100KM), 35000Ts (within 100-200KM), 70000Ts (200KM-360KM), 80000Ts (over 360KM).

The sending method in Embodiment 2 of the present invention includes the following steps:

Step 201, configure RACH according to the following method;

In the case of not changing the frequency domain resources, at this time, a cell of 360KM is to be covered. The transmission method of the present invention is: first calculate the size of N, at this time R=360Km, according to the calculation formula of N Get N=2, the sequence length is 2*24576Ts, the CP length is 24576Ts, the round-trip path delay reaches 2.412ms, and the GT length is 2.412ms, then the total length of RACH is 3*24576Ts+2.412ms=2.4ms+2.412ms=4.812 ms, only 5 TTIs are required in total, and since 5 TTIs are occupied in total, the GT length is adjusted to 2.6ms. A schematic diagram of the RACH preamble structure covered by 360KM in the embodiment of the present invention is shown in FIG. 8 .

To cover a 360KM cell without changing the frequency domain resources, when the traditional design method and detection algorithm are used, the CP and GT parts need at least 4.824ms, and the sequence part needs to stretch the 839RACH sequence change sampling to at least 2.4ms. It takes 7.224ms and takes up 8 TTIs. However, by adopting the sending method and detection method of the present invention, only 5 TTIs are needed, which effectively saves system time domain resources.

Step 202, obtain a new RACH preamble, and send the new RACH preamble according to the RACH configuration mode.

The receiving device for improving coverage capability of the present invention includes the following steps:

Step 401, peak detection sequence extraction: the base station side receives the random access signal, obtains the time domain position of the preamble according to the preamble configuration index, and extracts (N+2)*24576Ts long data including CP, sequence and part of GT , since N=2 in the present embodiment 2, then extract 4 sequences with a length of 24576Ts, and divide the data into 4 sections from front to back;

Step 402, performing parallel peak detection: performing parallel peak detection on the four peak detection sequences respectively. Record each detected preamble ID, time offset and calculate the SINR in the detection window.

Embodiment 1 has described step 402 in detail, and details are not repeated here.

Step 403, obtain initial time offset: obtain the initial time offset value through the joint judgment of the preamble ID result and SINR of each segment of parallel detection. Regardless of the detection results of paragraph 4: if the same leading ID is not only detected in paragraph 3, that is, it appears in paragraph 3 and at least one of paragraph 1 or paragraph 2, then the third paragraph is estimated If the same leading ID only appears in the third segment, compare the SINR of the third segment and the fourth segment, and take the estimated time offset value of the sequence with a larger SINR as the initial time offset value .

In this embodiment, a total of 4 user path delays are 1000Ts (within 100KM), 35000Ts (within 100-200KM), 70000Ts (200KM-360KM), and 80000Ts (over 360KM). As shown in Figure 9, only looking at the detection results of the first, second and third paragraphs, user 1 (with a delay of 1000Ts) falls within [0-24576) Ts, and the first, second and third paragraphs are all The leading ID0 of the user is detected, and the time offset value estimated in the third paragraph is used as the initial time offset value, and the time offset value is 1000Ts; The leading ID1 of the user is detected in both the 2nd and 3rd paragraphs, and the time offset value estimated in the 3rd paragraph is used as the initial time offset value, and the time offset value is 10424Ts; user 3 (with a delay of 70000Ts) falls in [2*24576Ts ~3*24576) Ts, only the leading ID2 of the user is detected in the third segment, compare the SINR of the third segment and the fourth segment, and the SINR of the fourth segment is larger, take the time offset value estimated in the fourth segment as the initial time Offset value, the time offset value is 20848Ts; user 4 (time delay is 80000Ts) exceeds 3*24576Ts, and the leading ID3 does not appear in the first, second, and third peak detection, and the time deviation of user 4 is too large to exceed the preamble format Out of range, user 4 fails in random access.

Step 404, perform calibration: by judging the appearance of the same preamble ID in N+1 parallel detection windows, perform different time offset calibrations respectively to obtain the correct time offset.

In this embodiment, regardless of the detection results of the fourth section, when the leading ID of user 1 appears in the three parallel peak detections of the first section, the second section, and the third section, the initial time offset value does not need to be calibrated , the initial time offset value is used as the last reported time offset value, that is, 1000TS is used as the last time offset value; the leading ID of user 2 does not appear in the first peak detection, but in the remaining second and third peak detections When it appears, add 24576Ts to the initial time offset value to get a calibration value of 35000TS, and use the calibration value as the last reported time offset value; When peak detection occurs in the third section, add 2*24576Ts to the initial time offset value to obtain a calibration value of 70000Ts, and use the calibration value as the last reported time offset value; the leading ID of user 4 does not appear in the first three peak detections When the time delay is too large, the random access fails or the preamble ID is not sent. Therefore, in the case of N=2, the maximum time offset range that can be detected by using this system is (2+1)*24576Ts=73728Ts, and the coverage radius is 360KM.

Embodiment 3 of the present invention:

Under the condition of not changing the LTERACH frequency domain resource, the sending method and the receiving method of the present invention are described by taking the coverage of 100KM cell as an example.

The sending method in Embodiment 3 of the present invention includes the following steps:

Step 201, configure RACH according to the following method;

In the case of not changing the frequency domain resources, at this time to cover a cell of 100KM, the transmission method of the present invention is: first calculate the size of N, at this time R=100Km, according to the calculation formula of N Get N=0, because N≥1, so N takes 1, the sequence length is 24576Ts, the CP length is 24576Ts, the round-trip path delay reaches 0.67ms, and the GT length takes 0.67ms, then the total length of RACH is 2*24576Ts+0.67ms =1.6ms+0.67ms=2.27ms, a total of 3 TTIs are needed, since a total of 3 TTIs are occupied, the GT length is adjusted to 1.4ms, so the coverage format of 100KM and the 200KM coverage format in Embodiment 1 of the present invention The same, as shown in Figure 6.

The time-frequency resource overhead required by this preamble format is the same as that of the existing LTE preamble format 3 covering 100KM. At the same time, there is no need to change the frequency domain position, only the RACH signal in the time domain is changed. The protocol changes are small, and it is convenient to change and easy to implement. .

The RACH preamble format supporting 100KM designed by the present invention is relatively flexible, and the receiving method of the present invention (described in detail in Embodiment 1, which will not be described here) can be adopted. The receiving method of the present invention not only realizes 100KM coverage but also can 200KM coverage is achieved, which effectively expands the coverage of the system; traditional receiving methods can also be used, which can achieve the same 100KM coverage as the existing RACH format 3. For this embodiment, because it only needs to cover 100KM, the traditional detection method can be used.

It can be seen from the above technical solutions that a transmission method for improving coverage provided by the present invention is to design both CP and T _SEQ as a sequence or sequence repetition in units of 24576Ts. Compared with traditional transmission, the RACH preamble of the present invention The occupied time domain resource is smaller. In the present invention, the CP is designed with a fixed length of 24576Ts. When the sequence part is N 24576Ts, the receiving end uses the characteristics of CP and T _SEQ sequence repetition to perform preamble detection and time offset calibration, which can support (N+1)*24576Ts The time offset range that the system can support is expanded from 24576Ts limited by the length of T _SEQ to (N+1)*24576Ts.

Under the sending method of the present invention, the receiving method for improving coverage provided by the present invention divides the (N+1)*24576Ts long sequence including CP into N+1 processing sequences with 24576Ts as the unit, and divides them into Perform peak detection, obtain the initial time offset value through the joint judgment of the preamble ID and SINR of each peak detection, and calibrate the initial time offset value. Due to the use of the CP length of 24576Ts and the characteristics of T _SEQ 24576Ts repetition, the system The maximum time offset range that can be supported is expanded from the traditional 24576Ts to (N+1)*24576Ts.

Without increasing frequency domain resources, the sending and receiving method provided by the present invention can achieve the purpose of expanding system coverage by occupying less time domain resources, requires less protocol modification, reduces overhead, and is easy to implement.

Although the present invention has been described in detail above and examples have been given, the present invention and applicable embodiments are not limited thereto, and those skilled in the art can make various modifications according to the principle of the present invention. Therefore, any modifications made according to the principles of the present invention should be understood as falling within the protection scope of the present invention.

Claims (7)

1. A kind of 1. accidental access signal method of sending and receiving applied to large coverage, it is characterised in that：Including following step Suddenly：

   User device configuration RACH RACH forms, the CP length of RACH forms, which is set, includes M repeatedly 24576Ts、T_SEQLength is arranged to the 24576Ts of N number of repetition, wherein M=1, N >=1；N is by formulaR is covering scene radius, and C is the light velocity, and Ts=1/ (15000*2048) s is obtained；By The protection interval GT that RACH signals take length adjustment RACH signals is extremely more than or equal to trip path time delay；

   For RACH signals after user's setting send configuration to base station, base station extracts the peakvalue's checking sequence of N+2 sections 24576Ts length Row；N+2 section peakvalue's checking sequences are subjected to peakvalue's checking respectively；The targeting sequencing ID results and noise detected side by side to every section Bias when cascading judgement than SINR obtains initial；By judging that same targeting sequencing ID goes out in N+1 parallel detection window Existing situation, to it is initial when bias carry out calibration obtain correct targeting sequencing ID and when bias, accidental access signal receive into Work(.
2. 2. a kind of accidental access signal method of sending and receiving applied to large coverage according to claim 1, its It is characterised by：The peakvalue's checking sequential extraction procedures process is：From the beginning the sequence extracted is divided successively backward with 24576Ts length For the 1st section, the section sequence of the 2nd section of N+2 section, parallel peakvalue's checking is carried out respectively to section sequence, records every section of peak value The targeting sequencing ID of detection and when bias and N+1 sections and N+2 sections detection window in SINR.
3. 3. a kind of accidental access signal method of sending and receiving applied to large coverage according to claim 2, its It is characterised by：It is as follows to obtain process for bias when described initial：The testing result of the N+2 sections in section sequence is removed, when N+1 sections And at least one section in remaining N sections detects same preamble serial ID, then using N+1 section sequence estimations when bias as just The bias during beginning；The testing result of the N+2 sections in section sequence is removed, is occurred when same targeting sequencing ID in N+1 sections, then Compare the SINR of N+1 sections and N+2 sections, take that section of big sequence estimation of SINR come out when bias as it is initial when bias.
4. 4. a kind of accidental access signal method of sending and receiving applied to large coverage according to claim 1, its It is characterised by：The N+2 sections peakvalue's checking sequence carries out peak-value detection method respectively：Every section of peakvalue's checking sequence frequency displacement is arrived At DC, sequence is filtered and down-sampling, obtains the sequence that length is 2048Ts, carry out 2048 point quick Fourier conversion, 839 point original ZC sequences are therefrom extracted, locally the frequency domain sequence of female code is total to the multiple of RACH frequency domain sequences received with ZC sequences Yoke is multiplied, and carries out 1536 points of inverse discrete Fourier transform, obtains PDP time domain correlations；PDP time domain correlations are taken into average, The setting of threshold value and the detection of peak value are carried out according to PDP averages, records the every section of targeting sequencing detected ID and corresponding When bias, calculate obtain SINR.
5. 5. a kind of accidental access signal method of sending and receiving applied to large coverage according to claim 1, its It is characterised by：Bias calibration includes three kinds of modes when described：

   The first：Remove the testing result of N+2 sections, when same targeting sequencing ID in the parallel peakvalue's checking of preceding N+1 all During appearance, bias is not required to calibrate when initial, and bias is as the when bias finally reported when will be initial；

   Second：The testing result of N+2 sections is removed, when same targeting sequencing ID is at 1M (M=1N) When not occurring in individual peakvalue's checking, and occurring in remaining N+1-M peakvalue's checking, bias adds M* when will be initial 24576Ts obtains calibration value, using calibration value as the when bias finally reported；

   The third：The testing result of N+2 sections is removed, when same targeting sequencing ID does not occur in N+1 peakvalue's checking When, time delay is excessive to be caused random access failure or does not send out targeting sequencing ID.
6. 6. a kind of accidental access signal method of sending and receiving applied to large coverage according to claim 1, its It is characterised by：The SINR value is the PDP peak energies in detection window and total PDP energy ratios.
7. 7. a kind of accidental access signal method of sending and receiving applied to large coverage according to claim 6, its It is characterised by：The PDP peak energies are the energy that PDP peak maximums energy is nearby put plus peak value.