CN102592153A - RFID (radio frequency identification device) reverse signal receiving method for inhibiting system noise - Google Patents

Abstract

The invention discloses a RFID (radio frequency identification device) reverse signal receiving method for inhibiting system noise, which is suitable for card reader receiver design based on a passive RFID label system. According to the invention, a RFID card reader allocates the pigtail number and pigtail coefficient of a time domain pigtail filter on the front end of the receiver through the prior information of reverse signal in an emitter forward signal in real time, and the real-time allocated filter inhibits the system noise in the reverse signal. Meanwhile, the method adds difference information of signal envelope amplitude in the RFID reverse signal, and the information can be used by the receiver to judge and send 0, 1 data so as to further improve the demodulating and decoding success rate. The method solves the problems of low reading success rate and close reading distance of RFID card reader in the RFID system caused by the irresistible noise interference.

Description

A kind of RFID reverse signal method of reseptance that suppresses system noise

Technical field

The present invention relates to RF identification (RFID) system field, and be particularly related to RFID communication system based on the passive mode label.Be applicable to the design of reverse signal receiver.This method has solved in the rfid system because irresistible noise, cause the RFID card reader read success ratio low, read the near problem of distance.

Background technology

The increasing all trades and professions that are applied to of RFID communication system, especially based on the RFID communication system of passive mode label with its low-power consumption, low cost, be prone to dispose characteristic, rapidly occupied the main flow of RFID industry.The passive mode label is a kind of passive label equipment, and itself does not have the energy storage original paper, relies on the RFID card reader to send continual carrier wave (CW) signal to label and encourages as energy, starts the duty of label.Label through backscatter technique to the feedback reverse data of card reader, therefore, for the card reader receiver, its receive signal be CW signal, RFID label back-scattered signal and.Because the backscattered signal energy of label is very little, inevitably receives the influence of system noise.When RFID card reader and tag distances were nearer, the CW signal produced stronger energy incentive action to label, and this moment, label can the sufficient reverse signal of backscattering energy, and card reader is correctly received.But when the distance of card reader and label constantly increases, influenced by the forward path loss, the CW signal that label is received dies down gradually, and the energy of back-scattered signal also lowers thereupon.Simultaneously, because the reverse path loss increases, the reverse signal energy that the RFID card reader receives further reduces.When reverse signal was flooded by system noise, card reader and label can not be set up communication process.In the low-cost solution of RFID, system noise is an outwardness.Therefore concerning low-cost solution, the ability that system suppresses noise be improve the RFID card reader read distance and the key that reads success ratio.

Usually suppress system noise and mainly adopt frequency filter and two kinds of methods of over-sampling.The frequency filter method is the bandwidth according to the target restituted signal, and the designing filter band width carries out squelch to the frequency that target restituted signal bandwidth is outer.In the RFID reverse link communication, this width is about several MHz.Although this method can effectively suppress out-of-band noise, still residually in the spectral bandwidth of target restituted signal partly disturb, and this frequency domain filtering method is non-linear in time domain, can cause the distortion of signal time domain, promptly introduce new interference.Oversampler method is through the digital sampler far above signal bandwidth the RFID reverse signal to be carried out over-sampling, through curve fitting algorithm, falls sampling then, obtains level and smooth reverse signal curve.But this method is because the restriction of digital sample speed, and the over-sampling multiple that can provide is limited, makes that the gain that obtains in the practical application is not high.

In a word; Present various low-cost solution all is that system noise inhibition ability effectively all can not be provided; Make system noise seriously retrain the RFID card reader and read success ratio and read range index, thereby cause the application scenarios of the RFID communication system of passive mode label to have certain limitation.And through constantly promoting the radio circuit craft precision, the expensive solution of elimination system noise and the small configuration cost feature of RFID are conflicting, can't form main flow.

Summary of the invention

The problem that the present invention will solve is: existing low complex degree RFID receiver suppresses the method for system noise, or can not effectively suppress in-band noise, maybe enough demodulation gains can not be provided, and has certain limitation.Therefore, need a kind ofly consider, transmitter and receiver are mutually combined, adopt the receiver of low complex degree wave filter to suppress the system noise method for designing, read success ratio and read range index with the raising card reader from the whole aspect of RFID card reader.

Technical scheme of the present invention is: utilize in the transmitter forward signal prior imformation about reverse signal, the time domain tap filter of receiver control front end is realized the inhibition to system noise in the reverse signal.Then, the signal through wave filter is carried out demodulation, decoding.

Define reverse baseband signal f _BB(t):

f _BB(t)＝f _CW(t)+f _RFID(t)+n(t)

In the formula: f _CW(t) be CW ripple signal, f _RFID(t) be the RFID back-scattered signal, n (t) is a system noise, and n (t) N (0, σ ²), the energy of system noise is σ ²This signal is defined as through behind the digital sample:

f _BB(n)＝f _CW(n)+f _RFID(n)+n(n)

The present invention is based on the time domain tap filter, the mathematical model that defines this wave filter is h (n), n ∈ [0, N-1].Wherein N is a filter length, the coefficient of i time domain tap of h (i) expression wave filter, and this tap coefficient is normalized value, promptly

Therefore, as reverse baseband signal f _BB(t) through behind this wave filter, signal is:

$f_{\mathrm{BB}}^{\′}\left(n\right)=\underset{i=0}{\overset{L-1}{\mathrm{\Σ}}}{f}_{\mathrm{BB}}(n+i)\×h\left(i\right)$

$=f_{\mathrm{CW}}^{\′}\left(n\right)+f_{\mathrm{RFID}}^{\′}\left(n\right)+n^{\′}\left(n\right)$

Therefore, the reverse signal behind wave filter still can be divided into CW ripple signal, RFID reverse signal, noise signal three parts.According to the equivalent mathematical model of time domain filtering, the three can be expressed as successively:

$\left\{\begin{array}{c}{f}_{\mathrm{CW}}^{\′}\left(n\right)=\underset{i=0}{\overset{L-1}{\mathrm{\Σ}}}{f}_{\mathrm{CW}}(n+i)\×h\left(i\right)\\ f_{\mathrm{RFID}}^{\′}\left(n\right)=\underset{i=0}{\overset{L-1}{\mathrm{\Σ}}}{f}_{\mathrm{RFID}}(n+i)\×h\left(i\right)\\ n^{\′}\left(n\right)=\underset{i=0}{\overset{L-1}{\mathrm{\Σ}}}n(n+i)\×h\left(i\right)\end{array}\right.$

In the formula, CW ripple signal f _CW(n) become parameter slowly for the time in the ideal case, define it in RFID reverse link communication process, be constant A.Because h (i) is the normalization tap coefficient, can obtain f ' this moment _CW(n)=f _CW(n)=and A, therefore do not change through this wave filter CW ripple signal.For all the other two parts f ' _RFID(n) and n ' (n), through amplifying f ' _RFID(n), or/and, the purpose that n ' signal energy (n) can realize suppressing system noise dwindled.So this Design of Filter key is exactly through rationally controlling tap coefficient and time-delay length, realizing amplifying f simultaneously _RFID(n) and dwindle n ' signal energy (n).

Backscatter link speed (BLF) is based in the rfid system of passive mode label, the variable of definition reverse link rate.In real system, card reader makes label according to this BLF reflection reverse signal through the BLF of RFID forward link configuration label.That is to say that for the RFID card reader generally, the BLF value is a priori known information.Under the condition of known BLF, the RFID receiver is to know fully to the reverse signal waveform.Simultaneously, no matter be two-way space code coding (FM0), still Miller-2, Miller-4, Miller-8 are encoded, and all are to be made up of four kinds of waveforms among Fig. 3.Therefore, theoretical according to matched filter, the picture left above is as the reference model of time domain tap filter h (n) among selection Fig. 3, and this model still has tap number and two parameters of tap coefficient to confirm.

On the tap coefficient value, based on (n) energy criteria of minimization system noise n ', define this tap coefficient and adopt equivalent normalization value, i.e. the weights coefficient of each tap equals the inverse of total tap number.Therefore, the time domain expression formula that defines this wave filter is:

$h\left(n\right)=\frac{1}{N}$ (0＜n＜N)

Four kinds of waveforms among definition Fig. 3 are respectively:

$\left\{\begin{array}{c}{f}_{0+}\left(n\right)=H,(0<n<\frac{T}{2});L,(0<n<\frac{T}{2})\\ f_{0-}\left(n\right)=L,(0<n<\frac{T}{2});H,(0<n<\frac{T}{2}),\\ f_{1+}\left(n\right)=H,(0<n<T)\\ f_{1-}\left(n\right)=L,(0<n<T)\end{array}\right.$

Wherein, H, L, T represent high level amplitude, low level amplitude, the time-domain sampling cycle of this signal respectively, and the time-domain sampling cycle is the inverse of the said BLF of preamble.According to the coded system that is adopted in the rfid system, for example, two-way space code coding (FM0); Miller-2, Miller-4, Miller-8 coding; And be not limited to above-mentioned coded system, can draw as drawing a conclusion: at any adjacent data intersymbol, must the upset of generation amplitude.For example, for continuous data 00, only can be 0 ₊0 ₊Or (with) 0 _-0 _-, but do not allow for 0 ₊0 _-Or (with) 0 _-0 ₊Therefore, can reduce following table

Data	Can adopt waveform	Can not adopt waveform
			00	0 +0 +Or (with) 0 -0 -	0 +0 -Or (with) 0 -0 +
01	0 +1 +Or (with) 0 -1 -	0 +1 -Or (with) 0 -1 +
			10	1 +0 -Or (with) 1 -0 +	1 +0 +Or (with) 1 -0 -
11	1 +1 -Or (with) 1 -1 +	1 +1 +Or (with) 1 -1 -

Based on the constraint between above-mentioned waveform, successively to f ₀₊(n), f _0-(n), f ₁₊(n), f _1-(n) through time domain tap filter h (n), can draw four kinds of waveforms through wave filter, its synoptic diagram is corresponding diagram 4 respectively.At this moment, between above-mentioned 0 and 1 data waveform, except the difference on time domain, increased difference, just increased the dimension of judgement envelope amplitude, promoted the accuracy of judgement at envelope amplitude.Because filtered signal, the maximal value of its envelope amplitude is relevant with tap number N, and therefore, designing filter tap number of criteria is defined as, envelope amplitude difference between maximization 0 and 1 data waveform.According to mathematical derivation, the ultimate value of 0 and 1 data waveform is respectively

$\left\{\begin{array}{c}{0}_{\mathrm{MAX}}=L+\frac{(H-L)\&times;\mathrm{<figure-callout\; id="2"\; label="T"\; filenames="426907DEST\_PATH\_HSB00000528077700012.png,DEST\_PATH\_HSB00000498577300013.png"\; state="\{\{state\}\}">T</figure-callout>}}{2}\\ 0_{\mathrm{MIN}}=L+\frac{(H-L)\&times;(N-T)}{2}\\ 1_{\mathrm{MAX}}=L+(H-L)\&times;\mathrm{<figure-callout\; id="1"\; label="N"\; filenames="115346DEST\_PATH\_HSB00000498577300012.png,426907DEST\_PATH\_HSB00000528077700012.png"\; state="\{\{state\}\}">N</figure-callout>}\\ 1_{\mathrm{MIN}}=L\end{array}\right.$

In the formula: the definition of H, L, T, N is consistent with preamble.Criterion based on above-mentioned four the value differences of maximization can draw, and when N=2/3 * T, system obtains maximum waveform judgement difference.According to the reciprocal relation of T and BLF, can draw the optimum tap coefficient N=2/3 * T of wave filter, be that the nothing that can be used as receiver is tested information and known.Therefore, this invention possesses feasibility in practice.

Fig. 3 is RFID reverse signal f _BB(t) typical waveform, f in this waveform _CW(t) energy is f _RFID(t) and the 40dB of n (t), f _RFID(t) equal with the energy approximation of n (t).As can be seen from the figure, this moment, the RFID reverse signal received the obvious noise interference.Read success ratio, read range index and had a strong impact on.Fig. 4 is the time domain waveform behind this signal process wave filter of the present invention, and filter tap is 20.Through filtered signal, obviously become level and smooth, and increased the difference of signal envelope amplitude between RFID reverse signal part 0,1 data.

Description of drawings

Fig. 1 is the synoptic diagram of typical four kinds of oscillograms in the RFID reverse signal.

Fig. 2 is under the noise-free case, and the RFID reverse signal is through the waveform synoptic diagram behind the wave filter of the present invention, and Ben Tu is corresponding one by one with Fig. 1.

Fig. 3, last figure are RFID reverse signal typical waveform figure, and figure below is the enlarged diagram of useful signal part.

Fig. 4, last figure are that the signal of describing among Fig. 3 is through the waveform synoptic diagram behind the present invention.Figure below is the enlarged diagram of useful signal part.

Fig. 5 is that the Physical layer of card reader 1 realizes block diagram.

Fig. 6 is the realization synoptic diagram of wave filter 13.

Embodiment

Specify embodiment of the present invention below:

The present invention is applicable to rfid system card reader 1 device.Card reader 1 is carried out data interaction through reverse signal before wireless and label 2.In force, card reading system 1 can adopt field programmable logic array (FPLA) (FPGA) or digital signal processor DSP or central processing unit (CPU) or microprocessor (MCU) realization.At first, card reader is through protocol processes 3, generates the data mutual with RFID label 2.Then to these data process coding 4 and modulation 5 successively; Realize the FM0 coding with (or) the MILLER-2 coding with (or) the MILLER-4 coding with (or) the MILLER-8 coding; And DSB modulation with (or) the SSB modulation with (or) the PR-ASK modulation, obtain digital signal to be sent.This digital signal after the up-conversion 8, becomes radiofrequency signal through digital-to-analog converter 7, and promptly the RFID forward signal is initiated the communication to RFID label 2.

Label 2 carries out reverse link communication through backscattering mode and card reader 1.Card reader 2 is through down coversion 8, and A-D converter 9 obtains the reverse digital baseband signal of RFID, i.e. f described in the present invention _BB(n).Fig. 3 is the typical time domain waveform figure of this signal.

In order to realize the output signal to A-D converter 9, i.e. f _BB(n), the squelch effect, this signal is carried out wave filter 13 handles.This wave filter is the time domain tap filter, and is as shown in Figure 6.Wave filter comprises 6 parts altogether, is successively:

(1) the information interaction path of A-D converter 9 and wave filter 13, the sampling rate of main transmission mode transformation of variables.

(2) the information interaction path of reverse prior imformation 12 and wave filter 13, mainly transmit reverse signal BLF or (with) time value of symbol time length.Utilize in the value of being somebody's turn to do and (1), the sampling rate of A-D converter 9 feedbacks can obtain the parameter T among the present invention jointly.

(3) the information interaction path of theoretical value 14 and wave filter 13 mainly transmits in the present invention the theoretical derivation value of wave filter optimization length.In force, this value is set at 2/3.

(4) computing unit 15, utilize A-D converter 9, reverse prior imformation 12,14 feedacks of theoretical value, can obtain the optimal filter tap length N of the reverse RFID signal of target demodulation

(5) according in the computing unit 15, the optimal filter tap length N of calculating, the delay unit number of configuration wave filter is N.

(6) according in the computing unit 15, the optimal filter tap length N of calculating, all tap coefficients of configuration wave filter are 1/N.

Among Fig. 5, the information interaction relation of utilizing A-D converter 9, reverse prior imformation 12,14 3 modules of theoretical value and filter module 13 has been described.Signal through wave filter 18 is as shown in Figure 4, as can be seen from the figure, compares with Fig. 3 signal before wave filter 13, possesses following three aspect characteristics:

(1) signal envelope becomes smoothly, realizes squelch;

(2) kept in the original signal, a level upset of the inner generation of 0 signal, the characteristic of level upset does not take place in 1 signal inside;

(3) increased the difference of 0,1 signal envelope amplitude, 0 signal envelope amplitude is lower than 1 signal envelope amplitude.

According to above-mentioned three aspect characteristics, will carry out demodulator decorder 110 and demodulator decorder 215 through the signal of wave filter 13.Wherein demodulator decorder 110 is the same with code translator in the traditional scheme, only distinguishes 0,1 signal through the inner level upset of signal.Demodulator decorder 215 is to distinguish 0,1 signal according to the signal envelope amplitude.Demodulator decorder 110 merges through merge cells 16 with the decode results of demodulator decorder 215; In merging; Can give to make up arbitrarily to the two-way demodulating data, for example, use the result of demodulator decorder 110 separately; Use the result of demodulator decorder 215 separately, use demodulator decorder 110 and demodulator decorder 215 jointly.

Decode results is imported protocol processing unit 11 carry out the parsing of RFID reverse signal.

Through above-mentioned implementation method, can the present invention be deployed in the RFID card reader kind equipment.

Claims (8)

1. suppress the RFID reverse signal method of reseptance of system noise, it is characterized in that utilizing the time domain tap filter, the rfid system reverse signal is carried out filtering.The RFID reverse signal can increase the envelope amplitude characteristic quantity through after this filter process on the basis of envelope time domain waveform.Then, the reverse receiver of RFID divides other based on envelope time domain waveform, envelope amplitude filtered signal to be carried out demodulation coding.At last the demodulation coding result is merged, read success ratio, read distance to improve rfid system.

2. method according to claim 1, wherein said time domain tap filter is real-time configurable tap filter.This wave filter is through resolving reverse signal prior imformation, hardware A-D converter sampling rate, theoretical value; The configuration guidelines of the tap number of acquisition time domain tap filter and the coefficient of each tap is N=2/3 * T; N is the tap filter number, and T is a RFID reverse signal digital sample number of times.

3. method according to claim 2, wherein said reverse prior imformation is meant the time span of RFID reverse signal, this information can obtain in the forward direction agreement.Wherein said hardware A-D converter sampling rate is meant the sampling rate of reverse receiving system, this information can through hardware feedback with (or) acquisition artificially is set.Wherein said theoretical value is a constant, middle finger 2/3 of the present invention.

4. method according to claim 2, wherein said analytic method is meant N=2/3 * T, T is a RFID reverse signal digital sample number of times, obtains jointly through reverse prior imformation described in the claim 3 and hardware A-D converter.

5. method according to claim 2, the tap number of wherein said time domain tap filter is the Arithmetic Formula N as a result in the claim 4.

6. method according to claim 2, the tap coefficient of wherein said time domain tap filter is whole identical coefficient of equivalences.

7. method according to claim 1, the envelope amplitude characteristic quantity of wherein said increase are meant that maximum, the minimum envelop amplitude of data 0 and data 1 signal that adopts is different in this signal.Therefore receiver can distinguish data 0 and data 1

8. method according to claim 1, wherein said demodulation coding merging method as a result are meant, the result based on envelope time domain waveform, two kinds of demodulation coding methods of envelope amplitude is carried out combination in any.Comprise, select result, and abandon result based on the envelope amplitude method based on envelope time domain waveform method; Selection is based on the result of envelope amplitude method, and abandons the result based on envelope time domain waveform method; The result of two kinds of methods is compared, and unanimity is then exported any decode results, inconsistent then demodulation failure.

Images