CN101102147B - A Timing Advance Estimation Method Applicable to GSM/EDGE System - Google Patents

Abstract

The invention relates to the field of mobile communication, in particular to a method for estimating the timing advance suitable for the GSM/EDGE system. The method obtains an initial value of the timing advance by performing energy calculation and comparison in the channel, and can quickly find an accurate value of the timing advance according to the obtained initial value of the timing advance. It is applicable under various situations (the system is in a static or multipath environment, it is used for voice service or data service, it is for a terminal system or a base station system).

Description

A Timing Advance Estimation Method Applicable to GSM/EDGE System

technical field

The invention belongs to the field of mobile communication, in particular to a method for estimating a timing advance (TA-TimingAdvance) applicable to a GSM/EDGE system. the

Background technique

As the second generation mobile cellular communication system, GSM has been widely used all over the world. However, with the rapid development of mobile communication technology and the diversification of bearer services, people's demand for data services continues to increase. Therefore, manufacturers are developing new, faster mobile data communication technologies, the most typical of which are HCSD (High Rate Circuit Switched Data) and GPRS (General Packet Radio Service) technologies. The highest rate of HCSD can reach 57.6kbit/s. GPRS introduced hierarchical switching in the core network for the first time, becoming the only way for GSM to evolve to the third-generation network. It can provide users with a link with a maximum rate of 171kbit/s, although HSCSD and GPRS have improved Data transmission rate, but because it still uses GMSK (Gaussian minimum shift keying) modulation method, the transmission rate it provides is still far from meeting people's increasingly higher requirements for communication systems, so it is necessary Adopt more advanced communication and signal processing technology to further expand the capacity of the GSM system. In order to meet people's ever-increasing demand for high-speed data services and the needs of the entire communication market, and enhance the competitiveness of existing systems, TIA of the United States and ETSI of the European Telecommunications Standardization Institute jointly proposed a wireless access technology based on TDMA - EDGE (Enhanced Data Rates for GSM Revolution), providing high-speed data communication services in the existing GSM frequency band. the

In order to provide a higher data communication rate in the existing cellular system, EDGE introduces a multi-level digital modulation method - 8PSK modulation. For 8PSK modulation, each modulation symbol can be mapped to 3 bits, which can provide higher bit rate and spectral efficiency compared with GMSK modulation, but its implementation complexity does not increase much. The modulation method of GMSK used in the GSM system is also a part of the EDGE modulation method. The symbol rate of the two modulation modes is 271kbit/s, and the net bit rate of each time slot is 22.8kbit/s (GMSK) and 69.2kbit/s (8PSK) respectively. the

Regardless of the original GSM system or the new EDGE system, the signal sent by the transmitter must be affected by the multipath channel. Therefore, in order to overcome the intersymbol interference caused by the multipath environment, a channel equalizer is usually required at the receiving end. . According to the GSM protocol, the data to be sent at the sending end is sent according to a burst (Burst), and in the middle of each burst, a training sequence with a length of 26 is placed. In order to ensure the equalization performance of the equalizer, the receiving end generally requires an extremely accurate starting position of the training sequence, that is, an accurate timing advance (TA) is required at the receiving end. In the traditional method, the receiver uses a known training sequence to correlate with the received burst at each symbol position, and the symbol position corresponding to the maximum correlation value is the estimated TA. However, in the actual system, this method is greatly affected by the multipath environment, which leads to inaccurate TA calculation, and then affects the balance result of the system. the

In the U.S. patent "Timing Estimation for GSM Bursts Basedon Past History" with the patent number US6724837, the method used is to make the training sequence correlate with them for the first received N bursts; The respective TAs of the N bursts are obtained from the maximum value of the sequence, and the obtained N TAs are averaged; finally, the respective TAs of the subsequent M bursts are estimated on the basis of the obtained average value of the TAs. In this patent, in order to ensure the accuracy of the subsequent M burst time advances obtained, the number N of bursts used to calculate the initialization TA is required to be as large as possible, but this patent does not guarantee that it is used to calculate the initialization The accuracy of the TA for N bursts of TA values. Therefore, if the value of N is too large, the overall performance of the receiver cannot be guaranteed because the performance of the N bursts received in advance cannot be guaranteed. On the other hand, this patent requires to take the average value of the burst TAs received continuously. This method is feasible for voice services based on circuit switching; but for data services based on packet switching, due to the The received two adjacent bursts of data may not all come from the same transmitter, so this patent cannot be applied to services based on packet switching, especially for GSM systems with EDGE functions. lost its practical significance. the

In the European patent "Iterative Estimation of Timing in GSMBursts" corresponding to the patent number US6859491, a method for iteratively estimating the timing advance and performing data equalization is proposed. For each received burst, this patent first assumes that TA has an initial value, and then makes TA increase by 1 in each cycle until a certain number of times N is cycled; and in each cycle, according to the given On this basis, the known training sequence is correlated with the sequence at the corresponding position in the equalization result and the correlation value is recorded; finally, the maximum correlation The TA value corresponding to the value is the real required value. Compared with the above patent, this patent overcomes the shortcomings of the above patent, and it can be used not only for voice services based on circuit switching, but also for data services based on packet switching. However, this patent needs to demodulate the received signal N times when calculating TA. If N is too large, the calculation is too complicated, so it is difficult to apply it to the actual system. the

The disadvantage of the technical solution disclosed in the above patents is that it is not suitable for the current GSM/EDGE system, either the calculation complexity is too high, or the application is difficult. Therefore, it is necessary to find a more practical and flexible method for finding the timing advance. the

Contents of the invention

The technical problem to be solved by the present invention is to provide a more practical and faster method for finding the timing advance. the

For solving the problems of the technologies described above, the present invention provides a kind of timing advance estimation method applicable to GSM/EDGE system, comprising:

a) Perform symbol derotation according to the difference in the modulation mode of the received burst;

b) Estimate the channel parameter sequence using the derotated data;

c) Calculate the mean value of the estimated energy of the channel with a preset length;

d) Find the position of the diameter with energy greater than the average energy, denoted as L ₀ ~L _N ;

e) Determine whether L _N -L ₀ is less than K ₁ ×M and L ₀ >K ₂ ×M, if yes, set the initial value of the timing advance to L ₀ -K ₂ ×M; if not, set the initial value of the timing advance is L ₀ ;

Among them, M is the sampling multiple, K1 is the dispersion length of the channel, and K2 is a preset threshold;

f) Based on the initial value of the time advance set above as the starting point, calculate the energy sum of the continuous K ₁ root diameter for each double sampling; find the sampling point m corresponding to the maximum energy among the M energy sums; obtain an accurate time advance value Add m to the initial value of the previously set timing advance.

According to the method of the present invention, the initial value of the time advance is found at a relatively fast speed, and the accurate value of the time advance can be quickly found according to the obtained initial value of the time advance. The method has less computational complexity and can be used in each All of these situations (the system is in a static or multipath environment, used for voice services or data services, and for terminal systems or base station systems) are applicable. the

Description of drawings

Figure 1 shows the basic schematic diagram of the GSM/EDGE system transceiver. the

Fig. 2 is the data format of conventional burst pulse in GSM/EDGE system. the

Fig. 3 is a flow chart of timing advance estimation. the

Fig. 4 is a schematic diagram of NB single sampling channel estimation energy. the

Detailed ways

The implementation of the technical solution will be further described in detail below in conjunction with the accompanying drawings. The above and other objects, features and advantages of the present invention will be apparent to those skilled in the art from the description of the method of the present invention. the

Referring to Fig. 1, at the transmitting end, the data output by the source coding module 101 is sent into the channel coding interleaving module 102 and then sent into the wireless channel after passing through the baseband GMSK/8PSK modulation module 103, the up-conversion module 104 and the transmitting antenna 105; at the receiving end The signal received by the receiving antenna 106 is sent to the signal demodulation module 109 and the time advance and channel parameter estimation module 108 respectively through the down-conversion module 107, and the channel parameter and the time advance are estimated by the module 108 and then sent to the signal demodulation module 109 , then the signal demodulation module 109 demodulates the received signal according to the received information, the demodulation result is sent to the deinterleaving and decoding module 110 , and finally output through the information source decoding module 111 . the

Fig. 2 is the data format of conventional burst pulse in GSM/EDGE system. For the normal burst pulse NB (Normal Burst), its information data is divided into two groups of 58 symbols, of which 57 bits are data, and the other bit is frame stealing bit to indicate whether the data is user data or signaling. A 26-bit training sequence is inserted between these two pieces of data, in which 5 bits at the beginning and end of 26 bits are guard intervals, and the middle 16-bit data is used to estimate channel parameters and timing advance. Three tail bits "0" are added to both sides of the useful information segment. The last of NB data is 8.25bit time, without sending any signal, as the protection segment of adjacent time slots. the

With reference to Fig. 3, the estimation of the time advance of the present invention is carried out according to the following steps:

Step 301 , the receiving end modules 106 and 107 receive the burst and send it to the timing advance and channel parameter estimation module 108 . the

Step 302: Perform symbol derotation according to GMSK or 8PSK according to different modulation modes of the received bursts. the

Step 303, using the derotated data to estimate the channel parameter sequence. the

Step 304, calculate the mean value of the estimated channel energy of a certain length, the length ranges from 22 to 24; for NB and Access Burst (AB) due to the different search lengths, the length of the calculated energy mean value is different. the

Step 305, find the position of the path whose energy is greater than the average energy, denoted as L ₀ ˜L _N .

Step 306, judge whether L _N -L ₀ is less than K ₁ ×M and L ₀ >K ₂ ×M, if so, go to step 308, where M is the sampling multiple, K ₁ is the dispersion length of the channel, and K ₂ is a set The value range is 1-3; if not, go to step 307.

Step 307, set the initial value of TA to L ₀ .

Step 308, set the initial value of TA as L ₀ -K ₂ ×M.

Step 309, starting from the initial value of TA, calculating the energy sum of the continuous _K1 root diameters for each double sampling.

Step 310, find the sampling point m corresponding to the maximum energy among the M energy sums

Step 311, outputting the precise TA value plus m from the initial value of TA. the

Step 312, use the precise TA value to perform subsequent signal demodulation. the

FIG. 4 is a schematic diagram of channel estimation energy of a single-sampled NB, and the horizontal solid line in the figure indicates the average energy. Using the method of the present invention to estimate TA, then L ₀ =3, L _N =12, and when K ₂ =1, the final accurate value of TA is 2.

The timing advance estimation method of the present invention firstly obtains the initial value of TA by simply calculating the mean value of channel estimation energy, and then further obtains accurate TA estimation in sampling points on the basis of the obtained initial TA value. Compared with existing methods, this method can complete higher-precision timing advance estimation with low computational complexity, regardless of whether the receiving system is in a static or multipath environment, whether it is used for voice services or data services. the

The foregoing description of the preferred embodiment is provided to enable any person skilled in the art to make or utilize the invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles described herein can be applied to various other embodiments without use of inventive step. Therefore, the present invention shall not be limited to the embodiments shown here, but shall be based on the widest scope consistent with the principles and novel features disclosed herein. the

Claims (4)

1. A time advance estimation method applicable to GSM/EDGE system is characterized in that, comprising: A. Perform symbol inverse rotation according to the difference in the modulation mode of the received burst; B. Estimate the channel parameter sequence by using the derotated data; C. Calculating the mean value of channel estimation energy with a preset length; D. Find the position of the diameter whose energy is greater than the average energy, denoted as L ₀ ~L _N ; E. Determine whether L _N -L ₀ is less than K ₁ ×M and L ₀ >K ₂ ×M, if yes, set the initial value of the timing advance to L ₀ -K ₂ ×M; if not, set the initial value of the timing advance is L ₀ ; Among them, M is the sampling multiple, K ₁ is the dispersion length of the channel, and K ₂ is a preset threshold; F. Based on the initial value of the time advance set above as the starting point, calculate the energy sum of the continuous K ₁ root diameter for each double sampling; find the sampling point m corresponding to the maximum energy among the M energy sums; obtain an accurate time advance value Add m to the initial value of the previously set timing advance. 2. The timing advance estimation method suitable for GSM/EDGE systems according to claim 1, characterized in that: the value range of the channel length in the step C is 22-24. 3. The timing advance estimation method suitable for GSM/EDGE system according to claim 1, characterized in that: the value range of the threshold K ₂ is 1-3. 4. The timing advance estimation method applicable to GSM/EDGE system as claimed in claim 1, characterized in that: the channel length in the step C is different lengths to the regular burst pulse and the access pulse respectively.

Images