CN101370224A

CN101370224A - Multi-antenna modulate type detection method for EDGE system

Info

Publication number: CN101370224A
Application number: CNA2007101452400A
Authority: CN
Inventors: 马岚; 熊岗; 单淑伟; 倪萌
Original assignee: ZTE Corp
Current assignee: ZTE Corp
Priority date: 2007-08-17
Filing date: 2007-08-17
Publication date: 2009-02-18

Abstract

The invention provides a method of multi-antenna detecting and modulating type of EDGE system, comprising the following steps: 1, for the first normal burst pulse in the uplink packet service data block, respectively calculating the composite noise variance of GMSK and composite noise variance of 8PSK of the multi-antenna; step 2, comparing the composite noise variance of GMSK with the composite noise variance of 8PSK, ensuring the modulating type of the first normal burst pulse according to the comparing result; and step 3, ensuring the modulating types of the next three normal burst pulses according to the modulating type of the first normal burst pulse. So, the invention greatly reduces the calculating complexity and ensures the performance of the baseband system under the condition of satisfying the mobile communication system protocol standard.

Description

Method for detecting modulation type by multiple antennas in EDGE system

Technical Field

The invention relates to the field of mobile communication, in particular to a method for detecting modulation types by multiple antennas, which is suitable for an EDGE system.

Background

GSM has found widespread use worldwide as a second generation mobile cellular communication system. However, with the development of mobile communication technology and diversification of services, the demand of people for data services is increasing. The GSM system uses GMSK (gaussian minimum shift keying) modulation, which is far from the wide area coverage of 384kbit/s data rate and the local area coverage of about 2Mbit/s data rate of third generation mobile communication systems. At present, ETSI (european telecommunications standards institute) has decided to develop edge (enhanced Data rates for GSM evolution), which is a GSM evolution scheme for enhanced Data rates, as a future GSM evolution direction.

In order to provide higher data communication rates in existing cellular systems, EDGE has introduced a multi-level digital modulation scheme-8 PSK modulation. Since 8PSK modulation is a linear modulation, 3 consecutive bits are mapped to one symbol of I/Q coordinates, thereby providing higher bit rate and spectral efficiency. Meanwhile, the GMSK modulation scheme used in the GSM system is also a part of the EDGE modulation scheme. The symbol rates of both modulation schemes are 271kbit/s, and the net bit rates per slot are 22.8kbit/s (GMSK) and 69.2kbit/s (8PSK), respectively. Under different channel conditions, the EDGE system can provide 9 different modulation and Coding schemes — MCS (modulation and Coding scheme), wherein MCS 1-4 still use GMSK modulation scheme in the GSM system, and MCS 5-9 use 8PSK modulation scheme.

In an EDGE system, a packet traffic channel (PDTCH) carries user data in a packet-switched mode. All packet channels are based on a multiframe structure consisting of 52 TDMA frames, containing 12 data blocks, each consisting of 4 consecutive TDMA frames. The base station does not know whether the modulation scheme used by the block of packet service data transmitted by the terminal equipment is GMSK modulation or 8PSK modulation. The existing method is to use a method for judging the noise variance, that is, to calculate the noise variance by using a GMSK symbol rotation method, and then calculate the noise variance by using an 8PSK symbol rotation method, and compare the noise variance with the GMSK symbol rotation method to judge the modulation types of GMSK and 8 PSK. The advantage of this approach is that the complexity is relatively low, and in practical applications, the accuracy of NB demodulation in each uplink can be better guaranteed. However, this method has a disadvantage that when the signal is small, the modulation type is easily determined incorrectly, which may result in complete failure of demodulation.

In the "Detection of Modulation Type" patent application published under number WO2005/109808, a method of determining GMSK and 8PSK Modulation types in an EDGE system is disclosed. It uses the correlation of the I part and Q part of the received signal to calculate a metric reflecting the interference energy, and determines whether the GMSK modulation type or the 8PSK modulation type from the metric. The technology adopts Whittle-Wiggins-Robinson algorithm loop iteration to carry out white noise processing, and calculates the interference energy measurement of GMSK and 8 PSK. However, due to the adoption of the loop iteration algorithm, the complexity is very high, and the loop iteration algorithm is difficult to adopt in an actual system.

In U.S. patent application publication No. 2006/0215789 of Detection of signaling modulation Format Type, a method of determining GMSK and 8PSK modulation types in an EDGE system is disclosed. When judging the modulation type of the current pulse, the method needs to judge together according to the detection results of the modulation types of the previous pulse or the previous pulses on the same block, wherein the judgment is carried out by using the signal-to-noise ratio or noise energy information of the previous pulse. Compared with the traditional method for independently detecting the modulation type by a single pulse, the method has the advantage that the performance is improved by 3-5 dB. However, when detecting the modulation type of the current pulse, it is necessary to perform comprehensive judgment based on the previous result, and thus erroneous judgment and accumulation may occur. In addition, the judgment of the latter pulse modulation types depends on the judgment of the first pulse debugging type. Therefore, the success rate of the first pulse modulation type is not fundamentally solved.

In the "receiver for determining modulation type" of patent 02803159.8, a method of determining the modulation of a transmitted signal is disclosed. For the transmitted signal, estimating a channel for each demodulated signal, the channel estimation value comprising m taps, selecting n taps with the largest energy in each channel estimation tap, wherein n < m; the variance of each demodulated signal is estimated by the least square method based on the n taps, and the modulation scheme of the transmitted signal is determined by comparing the variances. It is mainly applied to the determination of GMSK and 8PSK signals. However, this method first demodulates the signal twice and then determines the type of modulated signal based on the calculated variance. The main disadvantage is that the judgment of GMSK and 8PSK modulation types is only carried out by using a single antenna, and the gain of multiple antennas is not fully utilized.

In summary, the prior art has the disadvantages that the advantages of multiple antennas are not fully utilized, the multiple antennas are easily affected by multiple channel interference, and the complexity is high. There is a need for a simple and effective method for detecting GMSK and 8PSK modulation types, which can not only ensure correct demodulation of uplink data, but also reduce the complexity of operation of the baseband system. At the same time, GMSK and 8PSK modulation types can be detected in a simpler and more efficient manner for EDGE systems. The noise variance calculated by multiple antennas is used for distinguishing GMSK modulation types and 8PSK modulation types in the uplink packet service data block, so that the performance of a baseband system can be guaranteed, and the operation complexity can be reduced.

Disclosure of Invention

In view of the above problems, an object of the present invention is to provide a method for detecting modulation types by multiple antennas in an EDGE system, which greatly reduces the computational complexity and ensures the performance of a baseband system on the premise of meeting the protocol specification of a mobile communication system.

An aspect of the present invention provides a method for multi-antenna detection of modulation types for an EDGE system, comprising the steps of: step 1, respectively calculating the combined noise variance of GMSK and the combined noise variance of 8PSK of multiple antennas for the first conventional burst in the uplink packet service data block; step 2, comparing the GMSK synthetic noise variance with the 8PSK synthetic noise variance, and determining the modulation type of the first conventional burst pulse according to the comparison result; and step 3, determining the modulation types of the following three regular bursts according to the modulation type of the first regular burst.

The step 1 includes: calculating a received signal strength indication value on each antenna in the multiple antennas respectively; and judging whether the received signal strength indicating value is greater than a preset threshold value, if so, respectively carrying out GMSK processing and 8PSK processing on the signal on the antenna corresponding to the received signal strength indicating value, otherwise, ignoring the signal on the antenna.

The GMSK treatment described above includes: and carrying out-pi/2 symbol flipping processing on the signals on the antenna and carrying out channel parameter estimation on the signals after the flipping processing.

When the GMSK processing is performed, channel parameter estimation is performed by a sliding correlation method, and the estimation is calculated by the following formula:

wherein,

is an estimate of the channel parameters at GMSK processing,

is the signal after flip processing, k is in the range of [0, M-1]M is the actual channel dispersion length, a (k) is the training sequence, and L_hThe length required for estimating the channel parameters is smaller than the length of the training sequence.

The 8PSK processing includes: and carrying out-3 pi/8 symbol flipping processing on the signals on the antenna, and carrying out channel parameter estimation on the signals after the flipping processing.

When 8PSK processing is performed, channel parameter estimation is performed by a sliding correlation method, and calculation is performed by the following formula:

wherein,

is an estimate of the channel parameters at GMSK processing,is the signal after flip processing, k is in the range of [0, M-1]M is the actual channel dispersion length, a (k) is the training sequence, and L_hThe length required for estimating the channel parameters is smaller than the length of the training sequence.

Wherein the GMSK synthetic noise variance is calculated by the following formula:

wherein N is the number of antennas, N_σThe number of training sequences for the computation of the variance of the synthetic noise,

in order to flip the processed signal, the signal is,

is the estimated channel parameters, and a (k) is the training sequence.

And, the composite noise variance of 8PSK is calculated by the following formula:

in order to flip the processed signal, the signal is,is the estimated channel parameters, and a (k) is the training sequence.

Further, in step 2, when the 8PSK synthesis noise variance is larger than the GMSK synthesis noise variance, the method includes the following processes: and determining the modulation type of the first conventional burst as GMSK, performing GMSK demodulation, and setting Demod _ Flag to GMSK.

In step 2, in the case where the composite noise variance of 8PSK is not greater than the composite noise variance of GMSK, the following processing is included: the modulation type of the first conventional burst is determined to be 8PSK, 8PSK demodulation is performed, and Demod _ Flag is set to 8 PSK.

The following processing is included in step 3: after the first conventional burst demodulation is completed, determining the modulation types of the following three conventional bursts according to Demod _ Flag, and demodulating the following three conventional bursts according to the modulation types.

Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

Drawings

The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:

fig. 1 is a flow chart of a method of multi-antenna detection of modulation types for an EDGE system in accordance with the present invention;

fig. 2 is a basic diagram of a channel model of a mobile communication system;

fig. 3 is a diagram of a data format of a conventional burst in an EDGE system;

fig. 4 is a flow diagram of multiple antenna detection of GMSK and 8PSK modulation types in accordance with an embodiment of the present invention;

fig. 5 is a schematic diagram of a performance curve under TU50 by using a method of detecting GMSK and 8PSK modulation types using multiple antennas when an actual transmission modulation type is GMSK according to an embodiment of the present invention; and

fig. 6 is a schematic diagram of a performance curve under TU50 by using a method of multiple antennas for detecting GMSK and 8PSK modulation types when an actual transmission modulation type is 8PSK, according to an embodiment of the present invention.

Detailed Description

The preferred embodiments of the present invention will be described in conjunction with the accompanying drawings, and it will be understood that they are described herein for the purpose of illustration and explanation and not limitation.

Fig. 1 is a flowchart of a method of multi-antenna detection of modulation types for an EDGE system according to the present invention. As shown in fig. 1, the method for detecting modulation type by multiple antennas of EDGE system comprises the following steps: step S102, for the first conventional burst pulse in the uplink packet service data block, respectively calculating the composite noise variance of GMSK of multiple antennas and the composite noise variance of 8 PSK; step S104, comparing the GMSK synthetic noise variance with the 8PSK synthetic noise variance, and determining the modulation type of the first conventional burst pulse according to the comparison result; and step S106, determining the modulation types of the following three regular bursts according to the modulation type of the first regular burst.

The step S102 includes: calculating a received signal strength indication value on each antenna in the multiple antennas respectively; and judging whether the received signal strength indicating value is greater than a preset threshold value, if so, respectively carrying out GMSK processing and 8PSK processing on the signal on the antenna corresponding to the received signal strength indicating value, otherwise, ignoring the signal on the antenna.

wherein,

is an estimate of the channel parameters at GMSK processing,

wherein,

is an estimate of the channel parameters at GMSK processing,

in order to flip the processed signal, the signal is,

is the estimated channel parameters, and a (k) is the training sequence.

in order to flip the processed signal, the signal is,

is the estimated channel parameters, and a (k) is the training sequence.

Further, in step S104, when the composite noise variance of 8PSK is larger than that of GMSK, the following processing is included: and determining that the modulation type of the first conventional burst is GMSK, performing GMSK demodulation, and setting a demodulation Flag Demod _ Flag to GMSK.

In step S104, in the case where the composite noise variance of 8PSK is not greater than the composite noise variance of GMSK, the following processing is included: the modulation type of the first conventional burst is determined to be 8PSK, 8PSK demodulation is performed, and Demod _ Flag is set to 8 PSK. The following processing is included in step S106: after the first conventional burst demodulation is completed, determining the modulation types of the following three conventional bursts according to Demod _ Flag, and demodulating the following three conventional bursts according to the modulation types.

Specifically, fig. 2 is a basic diagram of a channel model of a mobile communication system. The baseband receiver receives data transmitted through a wireless channel port, the data firstly judges the modulation type of a received multi-antenna baseband I, Q signal through a modulation type module, then carries out diversity combination and demodulation according to the obtained modulation type, and the demodulated result is sent to a channel decoding module for channel decoding after being de-interleaved. For a packet traffic channel, one data block contains 4 consecutive NBs, and the demodulation results of the 4 NBs need to be output to a de-interleaving and decoding device.

Fig. 3 is a diagram of a data format of a conventional burst in an EDGE system. For NB, its information is divided into two groups of 58 symbols, where 57 bits are data and the other bit is frame stealing flag indicating whether this data is user data or signaling in CS traffic. A training sequence of 26 bits is inserted between the two pieces of data to estimate the channel parameters and the timing advance. The 3-bit "0" tail bits are added to both sides of the information segment. The last of the NB data has 8.25 bits of time and does not signal anything as a guard segment of the adjacent time slot. The NB may modulate the signal with GMSK or 8 PSK.

Fig. 4 is a flowchart of multi-antenna detection of GMSK and 8PSK modulation types according to an embodiment of the present invention. In an EDGE system, for a received signal, the signal model can be expressed as:

wherein,

for the received signal on the ith antenna, k is the position of a symbol point, wherein i is more than or equal to 1 and less than or equal to N, N is the number of antennas, and the actual system can be 1, 2, 4, etc. x (k) is the transmitted signal, h_i(k) For the channel parameter synthesized on the ith antenna, n_i(k) M is the actual channel dispersion length.

First, the RSSI estimation needs to be performed on the signal on each antenna in the first NB in the uplink packet service data block. The energy sum of the received signals can be calculated within a certain window range, and the RSSI value can be obtained by adopting a table look-up mode. If the RSSI value on this antenna is less than a certain Threshold, namely: RSSI < Threshold, then the signal for this antenna is negligible and no subsequent processing is performed. Where Threshold may be set to a fixed value, e.g., -120.

If the RSSI value on this antenna is greater than Threshold, the combined noise variance of GMSK and 8PSK, respectively, needs to be calculated. For the ith antenna, the symbol flipping process of GMSK is performed first, i.e. the I antenna is a reference signal

And after the overturning is finished, estimating the channel parameters. The channel parameter estimation may adopt a sliding correlation method, that is, a training sequence is used to perform sliding correlation to obtain the parameters of the channel. The formula is as follows:

wherein,

for the estimation of channel parameters in GMSK processing, k is in the range of [0, M-1 ]]M is the actual channel dispersion length, a (k) is the training sequence, and L_hThe length required for estimating the channel parameters is generally smaller than the length of the training sequence.

After the GMSK processing of the antenna is completed, 8PSK processing of the antenna is performed, and symbol inversion processing is also performed first, and the calculation formula is as follows:

the channel estimation is then performed as in the GMSK process. The estimated value of the channel parameter in the 8PSK processing is

The formula is as follows:

after the GMSK processing and 8PSK processing for all N antennas are completed, the combined noise variance calculation for GMSK and 8PSK is performed.

Wherein, the GMSK synthetic noise variance calculation formula is:

in order to flip the processed signal, the signal is,

is the estimated channel parameters, and a (k) is the training sequence.

The 8PSK composite noise variance calculation formula is:

in order to flip the processed signal, the signal is,

is the estimated channel parameters, and a (k) is the training sequence.

After the calculation of the variance of the synthetic noise of the multiple antennas is finished, the two are judged, and if Q is greater than the threshold, the two are judged_8PSK>Q_GMSKIf so, the modulation type of the current NB is judged to be GMSK, GMSK demodulation is carried out, and Demod _ Flag is set to be GMSK. Otherwise, 8PSK demodulation is performed, and Demod _ Flag is set to 8 PSK.

After the first NB in the uplink data block completes demodulation, the demodulation modes of the subsequent 3 NBs need to be determined according to the stored Demod _ Flag. If Demod _ Flag is equal to GMSK, then the subsequent 3 NBs all perform GMSK demodulation. Otherwise, the subsequent 3 NBs perform 8PSK demodulation. Finally, the results of 4 NB demodulation are deinterleaved and sent to a decoder.

Fig. 5 is a schematic diagram of a performance curve under TU50 by using a method of detecting GMSK and 8PSK modulation types using multiple antennas when an actual transmission modulation type is GMSK according to an embodiment of the present invention. As shown in fig. 5, when the modulation type detection is performed by using dual antennas, the probability of occurrence of misjudgment is much smaller than that of the conventional detection method using a single antenna. If the number of antennas N >2, the probability of a false positive will be smaller.

Fig. 6 is a schematic diagram of a performance curve under an urban model TU50(50 represents that the velocity of a mobile station is 50km/h) by using a method for detecting GMSK and 8PSK modulation types using multiple antennas when the actual transmission modulation type is 8PSK, according to an embodiment of the present invention. As in fig. 5, as shown in fig. 6, when the modulation type detection is performed by using dual antennas, the probability of occurrence of misjudgment is much smaller than that of the conventional detection method using a single antenna. If the number of antennas N >2, the probability of false positives is smaller.

In summary, the present invention provides a simple and effective method for detecting GMSK and 8PSK modulation types for EDGE systems. Namely, GMSK and 8PSK composite noise variance calculation is carried out on a first NB in an uplink packet service data block based on multiple antennas, and the modulation type is judged according to the composite noise variance. Meanwhile, the demodulation modes of other 3 NBs in the uplink data block are determined according to the modulation type in the first NB. On the premise of meeting the protocol specification of the mobile communication system, the method greatly reduces the calculation complexity and ensures the performance of a baseband system.

The above is only a preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes will occur to those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for multiple antenna detection of modulation types for EDGE systems, comprising the steps of:

step 1, respectively calculating the combined noise variance of GMSK and the combined noise variance of 8PSK of multiple antennas for the first conventional burst in the uplink packet service data block;

step 2, comparing the synthesized noise variance of the GMSK with the synthesized noise variance of the 8PSK, and determining the modulation type of the first conventional burst pulse according to the comparison result; and

and 3, determining the modulation types of the following three conventional burst pulses according to the modulation type of the first conventional burst pulse.

2. The method of claim 1, wherein step 1 comprises:

calculating a received signal strength indication value on each antenna of the multiple antennas respectively; and

and judging whether the received signal strength indicated value is greater than a preset threshold value, if so, respectively carrying out GMSK processing and 8PSK processing on the signal on the antenna corresponding to the received signal strength indicated value, and otherwise, ignoring the signal on the antenna.

3. The method of claim 2, wherein the GMSK processing comprises:

and carrying out-pi/2 symbol flipping processing on the signals on the antenna and carrying out channel parameter estimation on the signals after the flipping processing.

4. The method of claim 3, wherein the channel parameter estimation is performed by a sliding correlation method, and is calculated by the following formula:

wherein,

is an estimate of the channel parameters at GMSK processing,

5. The method of claim 2, wherein the 8PSK processing comprises:

and carrying out-3 pi/8 symbol flipping processing on the signals on the antenna, and carrying out channel parameter estimation on the signals after the flipping processing.

6. The method of claim 5, wherein the channel parameter estimation is performed by a sliding correlation method, and is calculated by the following formula:

wherein,

for the estimation of the channel parameters for 8PSK processing,

7. The method according to claim 4 or 6, wherein the GMSK composite noise variance is calculated by the following formula:

in order to flip the processed signal, the signal is,

is the estimated channel parameters, and a (k) is the training sequence.

8. The method of claim 4 or 6, wherein the 8PSK composite noise variance is calculated by the following equation:

in order to flip the processed signal, the signal is,

is the estimated channel parameters, and a (k) is the training sequence.

9. The method according to claim 1, wherein said step 2 comprises, in case that the 8PSK composite noise variance is larger than the GMSK composite noise variance, the following steps:

and determining that the modulation type of the first conventional burst is GMSK, performing GMSK demodulation, and setting a demodulation Flag Demod _ Flag to GMSK.

10. The method according to claim 1, wherein said step 2 comprises, in the case that the 8PSK composite noise variance is not greater than the GMSK composite noise variance, the steps of:

and determining that the modulation type of the first conventional burst is 8PSK, performing 8PSK demodulation, and setting Demod _ Flag to be 8 PSK.

11. The method according to claim 9 or 10, wherein the step 3 comprises the following processes:

and after the first conventional burst is demodulated, determining the modulation types of the subsequent three conventional bursts according to the Demod _ Flag, and demodulating the subsequent three conventional bursts according to the modulation types.