A method of reducing mobile communication system decoding delay
Technical field
The invention belongs to mobile communication technology field, especially satellite mobile communication receiving end Digital Signal Processings to lead
Domain is related to a kind of method for reducing mobile communication system decoding delay.
Background technique
Geostationary satellite mobile communication system (GMR, GEO-Mobile Radio Interface) is european telecommunication mark
The satellite that Zhun Hua association (ETSI, European Telecommunications Standards Institute) formulates is mobile
Communication standard.GMR standard is from ground digital honeycomb global mobile communication (GSM, Global System for Mobile
Communication) standard evolution, therefore satellite mobile terminal can access GSM core network.Since GMR is mobile logical
Letter system and gsm mobile communication system difference over the wireless channel, therefore the two has very big difference on a physical layer.With GSM
Mobile radio system to general packet radio service technology (GPRS, General Packet Radio Service) and
The evolution of third generation digital communication (3G, 3rd Generation) standard, the also evolution therewith of GMR standard, successively issue GMR-
1, GMR-GPRS and GMR-3G.GMR-1 corresponds to GSM standard, mainly supports basic circuit domain speech and facsimile service;GMR-
GMPRS corresponds to GPRS standard, mainly supports Packet data service;GMR-3G corresponds to 3G (WCDMA, Wideband Code
Division MultipleAccess) standard, mainly support high-speed packet data service.
In mobile communication system, it is contemplated that wireless channel and some enchancement factors influence, therefore are typically all according to nothing
Whole balanced, the interpretation method performance of line channel worst situation design receiving end.But under true wireless environment, especially extensively
It broadcasts under channel situation, base station end/gateway station is with full power broadcast wireless signal, for most of terminal, receives letter
Number power can be much larger than reference sensitivity, and signal quality also can be horizontal much higher than normal received signal.In addition GMR mobile communication system
System uses channel decoding technology, sends mistake in wireless channel transmission process to correct reception signal.In GMR satellite
In communication system, since geostationary satellite mainly uses wireless signal pass-through mode, which includes power amplifications etc. to operate.
If the mistake of longer duration, such as (burst) time interval of the burst more than 1 occur for wireless channel, it will cause whole
A receiving end receives declining to a great extent for digital signal quality.
For above-mentioned similar situation, if still using conventional receiver end Digital Signal Processing process, letter cannot be not only played
Road decoding action, and because sophisticated signal processing method in receiving end causes centainly to postpone.Therefore, guaranteeing to receive in receiving end
Under the premise of signal quality, how to reduce signal processing delay as far as possible is problem in need of consideration.
Summary of the invention
In view of this, the purpose of the present invention is to provide a kind of method for reducing mobile communication system decoding delay, the party
It is poor that method emphasis considers that wireless channel receives several burst signals, and major part receives the preferable situation of burst signal, if when receiving
Dry burst signal is poor prescribes a time limit lower than preset gate, and the filling signal for directlying adopt zero information is decoded.This method is guaranteeing to connect
On collection of letters quality base, takes full advantage of receiving end and deinterleave to the debunching action of error message and the error correction of channel decoding
Performance not only reduces and receives error burst data to the influence of receiving end performance, and it is complicated to reduce receiving end balancing procedure
Degree.For entire receiving terminal system, receiving end algorithm is not only increased to wireless environment adaptability, and greatly reduce
Signal processing time delay.
In order to achieve the above objectives, the invention provides the following technical scheme:
A method of reducing mobile communication system decoding delay, comprising the following steps: S1: estimation Signal to Noise Ratio (SNR) (SNR,
Signal to Noise Ratio);S2: it decides whether effectively to receive data;S3: balanced effectively to receive data;S4: filling nothing
Effect receives data;S5: it deinterleaves, channel decoding.
Further, in step sl, channel estimation, SNR estimation are carried out respectively according to every section of condition code, specifically include:
Channel estimation h:
It carries out channel with known condition code or the reception signal u estimated progress related operation using data r is received and estimates
Meter, it may be assumed that
H=xcorr (r, u)
Wherein, r indicates received wireless signal, the reception signal that u indicates known condition code or estimated, xcorr table
Show related operation;
Estimate SNR:
SNR is estimated using the logarithm of channel estimation power and noise power ratio, it may be assumed that
Wherein, h indicates radio channel estimation, i=1,2 ..., N;
Alternatively, estimating SNR using channel estimation power and noise power ratio, it may be assumed that
Wherein, h indicates radio channel estimation, i=1,2 ..., N.
Further, in step s 2, whether effectively in the following ways judgement receives data: it is predefined whether SNR is greater than
Thresholding takes SNR to be greater than the condition code left and right ends data of thresholding as effectively reception data.
Further, in step s 2, whether effectively in the following ways judgement receives data: to different characteristic estimation
SNR is ranked up, and takes the maximum several burst of SNR as effectively reception burst.
Further, in step s3, using decision feedback equalization or the equalization methods of maximum-likelihood sequence estimation, to process
What wireless channel disturbing factor influenced is that valid data carry out equilibrium by step S2 judgement, is made by wireless channel disturbing factor shadow
Loud reception signal maximal possible recovery sends effective information, provides necessary condition for subsequent decoding.
Further, in step s 4, it is filled using zero information to not normally receiving data: door is less than to all SNR
The burst of limit is filled;Alternatively, several burst the smallest to SNR carries out thresholding comparison, only to be less than the burst of thresholding into
Row filling.
Further, in step s 5, deinterleaving carries out receiving end error message sufficiently discrete, to guarantee channel decoding
To the error correction of isolated errors information;Due to most of SNR higher for receiving burst, pass through channel decoding (such as Viterbi
Decoding algorithm), filling channel receiving information can be recovered completely.
The beneficial effects of the present invention are: this method takes full advantage of reception on the basis of guaranteeing received signal quality
End is deinterleaved to the debunching action of error message and the error-correcting performance of channel decoding, is not only reduced and is received error burst data
Receiving end performance is influenced, and reduces receiving end balancing procedure complexity;For entire receiving terminal system, not only increase
Receiving end algorithm greatly reduces signal processing time delay to wireless environment adaptability.
Detailed description of the invention
In order to keep the purpose of the present invention, technical scheme and beneficial effects clearer, the present invention provides following attached drawing and carries out
Illustrate:
Fig. 1 is the flow chart of the method for the invention;
Fig. 2 is Normal Traffic9 channel data block interleaving, multiplexing structure figure;
Fig. 3 is Normal Traffic9 channel data block interleaving, multiplexing structure figure after deformation;
Fig. 4 is the simulation architecture figure of Normal Traffic9 channel;
Fig. 5 is the simulation result diagram that embodiment is realized.
Specific embodiment
Below in conjunction with attached drawing, a preferred embodiment of the present invention will be described in detail.
Fig. 1 is the flow chart of the method for the invention, as shown, the method for the invention specifically includes:
Step 1: estimation Signal to Noise Ratio (SNR) (SNR, Signal to Noise Ratio)
In GMR satellite mobile communication system, length be 9 burst Normal Traffic channel data block interleavings,
Multiplexing structure figure is as shown in Fig. 2, wherein G indicates protection interval, and u indicates condition code, and e indicates coded data.It can from Fig. 2
Out, condition code being unevenly distributed in single burst, but 4 sections of condition code approaches uniformities are distributed in 9 bursts, point
The received signal quality of different data section is not represented, as shown in Figure 3.In order to more preferably grasp 9 different burst reception signal qualities,
Every section of condition code can be carried out to channel estimation, SNR estimation respectively, respectively represent the reception signal of this section of condition code left and right ends
Quality, it can be considered former 9 bursts being reclassified as 4 new burst structures according to condition code.It then can be according to every
Section condition code carries out channel estimation, SNR estimation procedure respectively are as follows:
Channel estimation h:
The reception data r of jth section condition codejWith known condition code uj, channel estimation is carried out using correlation technique, it may be assumed that
hj=xcorr (rj,uj) (1)
Wherein, hjIndicate the channel estimation of jth section condition code, j=1,2,3,4, xcorr indicate related operation.
Estimate SNR:
Jth section condition code corresponds to SNRj, can be counted using the logarithm of channel estimation power and noise power ratio
It calculates, it may be assumed that
Preferably, SNR is estimated using channel estimation power and noise power ratio, it may be assumed that
Wherein, hjIndicate radio channel estimation, when j is respectively equal to 1,2,3,4, N is respectively equal to 12,6,6,6.Σ|hj|2
All channel estimation values that jth section condition code is estimated are overlapped by expression.
Step 2: deciding whether effectively to receive data
Preset threshold T first is the several times of noise, it may be assumed that
Wherein, γ indicates noise gate weighting coefficient, and based on experience value or simulation value determines.
Optionally, if SNRjGreater than predefined thresholding Tj, take this section of condition code left and right ends data as effectively reception number
According to otherwise as invalid burst reception data;
Preferably, the SNR to different sectionsjIt is ranked up, only as minimum SNRjGreater than thresholding TjWhen, take this section of condition code left side
Otherwise right two end datas are used as invalid burst reception data as data are effectively received.
Step 3: balanced valid data
Using decision feedback equalization or the equalization methods of maximum-likelihood sequence estimation, to by wireless channel disturbing factor shadow
Loud is that valid data carry out equilibrium by step 2 judgement, and maximal possible recovery sends effective information, provides necessity for subsequent decoding
Condition.
Step 4: filling invalid data
It is filled for invalid data using zero information on what is influenced by wireless channel a number of factors by step 2 judgement.
Wherein zero information just refers to, filling information does not include any effective reception information.Such as receive data area be [- 127,
It 127], then 0 is no any effective information for receiving data.It is filled using zero information, it on the one hand can be with
The reception data for reducing low SNR influence whole performance;On the other hand, it in information filling process, reduces introduce as far as possible
Interference information.
Step 5: deinterleaving, channel decoding
Deinterleaving is sufficiently discrete to receiving end error message progress, to guarantee that channel decoding entangles isolated errors information
Wrong function;It, completely can be with by channel decoding (such as Viterbi decoding algorithm) due to most of SNR higher for receiving burst
Unknown channel receiving information is recovered.Certain process may also include other some operations, such as be demultiplexed, go
Scrambling code and CRC decoding etc., depending on emulating the selection of link.
A kind of reduction mobile communication system time delay implementation method validity provided in order to further illustrate the present invention, below
It sends and receives emulation link to 9 channel tradition of Normal Traffic in GMR satellite communication system to verify, such as Fig. 4 institute
Show.Raw information completes transmitting terminal baseband procedure by CRC coding, convolutional encoding, intertexture, scrambling code, multiplexing and debugging, then passes through
Cross wireless channel.In receiving end, demodulated first, be then successively demultiplexed, go scrambling code, deinterleaving, folding coding and
CRC decoding, so as to calculate the bit error rate of entire emulation link by sending data and reception data.In order to simplify emulation
Process, using following simulated conditions:
1) wireless channel is white Gaussian noise, SNR=0,1,2 ..., 10dB;
2) demodulation is exported using hard decision, i.e., and 0,1, i.e. filling information is 0.5;
3) channel decoding uses the viterbi coding method of Hamming distance;
4) situation 1 is that tradition shown in Fig. 4 sends and receives process;The hypothesis of situation 2 meets all burst1 (as shown in Figure 2)
Data are received to be filled;Situation 3 assumes that reception data (as shown in Figure 2) to all burst1 and burst4 are filled;Situation 4
It is assumed that reception data (as shown in Figure 2) to all burst1, burst4 and burst6 are filled.
When noise gate weighting coefficient γ=4, fills burst number and be not more than 3, then performance loss is in receivable range
It is interior.The above method is used, is to be not required to for invalid information although increasing the estimation complicated degree of analysis of wireless channel
Carry out the relevant operations such as wireless channel equalization (40-60% for accounting for entire receiving end computation complexity).Therefore this hair is used
Bright method not only reduces entire receiving end signal processing complexity, and reduces receiving end signal processing delay.Fig. 5 is imitative
True result figure.
Finally, it is stated that preferred embodiment above is only used to illustrate the technical scheme of the present invention and not to limit it, although logical
It crosses above preferred embodiment the present invention is described in detail, however, those skilled in the art should understand that, can be
Various changes are made to it in form and in details, without departing from claims of the present invention limited range.