RU2416107C2 - Method of increasing range of radio frequency identification system on surface acoustic waves - Google Patents

Abstract

FIELD: physics.

SUBSTANCE: in the method of increasing the range of a radio frequency identification system, comprising use of a binary code to obtain code sequences used to encode identifier data, according to the invention, the code sequences selected differ from each other by not less than a given number of bits greater than one. The code sequences are generated either via enumeration of codes on a computer or using block codes or using a convolution code generator.

EFFECT: increase in range of a radio frequency identification system owing to encoding of identifier data with selected code sequences which differ by more than one bit.

3 dwg

Description

The invention relates to radio communications and can be used in systems of radio frequency identification on surface acoustic waves (SAWs).

Known methods of encoding data, for example, according to the patent of the Russian Federation 2344437, in which M delay lines (LZ) are used to encode data. When encoding data with one of the sets of M LZ, two adjacent codes differ by one bit. The identification range in this case is determined by the signal-to-noise ratio and the probability of error between adjacent codes for this signal-to-noise ratio. If the codes differ by one bit, a larger signal to noise ratio is required, and therefore a smaller distance to the mark is required to ensure this ratio.

The closest in technical essence to the proposed one is a method of encoding data by the on-off pulse method presented in the book by V. L. Dshkhunyan. Electronic identification. Contactless electronic identifiers and smart cards. / V.L. Dshkhunyan, V.F. Shangin. - M .: ACT Publishing House LLC: NT-Press Publishing House, 2004. - 695 pp., Ill., P. 251, adopted as a prototype.

The prototype method consists in encoding data on-off pulse, in which each possible position of the pulse encodes one bit of data for any binary data code.

In identifiers on SAWs (SAW transponders), the method of encoding data by turning on / off a pulse is widespread, in which each possible position of the pulse encodes one bit of data. A typical configuration of the SAW transponder of a radio frequency identification system (RFID) is shown in Fig.5.5 [2].

The presence or absence of a pulse in the response signal of the surfactant of the transponder is determined by the topology of the arrangement of reflectors on the piezocrystalline substrate. Each reflector creates its own pulse in the response signal of the SAW transponder, while the delay time between individual pulses is proportional to the spatial distance between the reflectors on the substrate. Therefore, by arranging the reflectors on the substrate accordingly, it is possible to generate the required binary code represented by the pulse train of the transponder response signal.

In [1] on pages 242-258 a radio frequency identification (RFID) system based on passive identifiers using SAW technology is described. Passive identifiers are crystals with a set of reflective structures. The identifier code consists of a unique set of reflectors, which corresponds to a unique sequence of electrical pulses generated as a result of the direct and inverse piezoelectric effect at the input / output of the identifier when an input electrical pulse is applied to it.

In the prototype method of data encoding carried out in a radio frequency identification system, the identifier code can be from 1 to 2 ^N , where N is the number of coding bits.

The range of the RFID system is determined by the formula:

where λ is the electromagnetic wavelength; P ₀ is the power sent by the reader; G _C and G _AND - gain (transmission) of the reader and antenna identifier, respectively; A _surfactant is the insertion attenuation of the identifier, kT ₀ BF is the significant thermal noise power (k is the Boltzmann constant, T ₀ is the absolute temperature, V is the system bandwidth, F is the noise figure),

SNR is the minimum signal-to-noise ratio required to reliably detect a received signal with a nominal design value or error probability.

This formula assumes that for the operation of the RFID system, it is necessary to provide a sufficient level of power of the reflected signal at the receiving input. However, when considering the operation of the allocation block of the received code, it turns out that the range of the RFID system depends, inter alia, on the structure of the code.

Let us consider in more detail the factors affecting the range of the RFID system. For the RFID system, the following structure will be the optimal receiver [2], Fig. 5.1.9., P.208.

Figure 1 presents a diagram of the optimal receiving device identifier codes.

Here s ₀ ... s _m are possible signals of codes from 0 to 2 ^N. A code equal to 0 corresponds to a situation when there is no signal from the identifier. E ₀ ... E _m is the energy of the corresponding signal. The maximum is selected by the strobe T, where T is the strobe period equal to the duration of the received signal.

Thus, to find the code, it is necessary to solve the system of equations

,

,

where n (t) is the noise.

Opening A _j , we obtain:

where R _ij is the cross-correlation coefficient for s _i and s _j signal;

R _nj - Gaussian random variable with zero mean and variance

, где N₀ - спектральная плотность шума.

where N ₀ is the spectral density of the noise.

.Then A _j will be a Gaussian random variable with an average value of R _ij - 0,5E _j and dispersion

.

Since the signal consists of an N-bit code, its energy can vary from 0 to ε ₁ · N, where ε ₁ is the energy of one bit. The value of R _ij depends on the number of common bits in s _i and s _j signals and can be from 0 to N.

The range of the RFID system is determined by the probability of correctly identifying the identifier code. To calculate this probability, consider the reception algorithm. At the input of the optimal receiver, we have the signal z (t) = n (t) + s _i (t). Using this signal, we calculate the values of A _j . A _j will be Gaussian random variables with the same variance and average values depending on i and j. A recognition error of the received code s _i occurs when any of A _{j is} greater than A _i . Based on the probability distributions A _j, we can calculate the probability that A _i will be greater than any of A _j .

The average value for A _i will be equal to: A _i = R _ij -0.5E _i = E _i -0.5E _i = 0.5 E _i .

The signal s _i (t) consists of m-significant bits, then E _i = m · ε ₁ . The average value of A _i will be 0.5 · mε ₁ . The closest codes to s _i (t) will be codes that differ by one significant bit.

For a signal s _i-1 (t) with one less significant bit, R _ii-1 = (m-1) · ε ₁ , E _i-1 = (m-1) · ε ₁ . The average value of R _ii-1 -0.5E _i-1 = 0.5 (m-1) · ε ₁ .

For a signal s _{i + 1} (t) with a large one significant bit, R _{ii + 1} = m · ε ₁ , E _{i + 1} = (m + 1) · ε ₁ . The average value of R _{ii + 1} -0.5E _{i + 1} = 0.5 (m-1) · ε ₁ .

As you can see, both nearest signals have the same average and variance.

, The probability density for A _i will be

,

for A _i-1 :

,

,

for A _{i + 1} : p _i-1 = p _{i + 1} .

Graphs of the probability density distribution are shown in figure 2.

.The probability of correct identification will be equal to the probability that z _i will be more than 0.5mε ₁ -0.25 ε ₁ , i.e.

.

и больше вероятности Р_вu1.If the codes differ by k significant, the probability of correct identification will be

and more probability P _vu1 .

Thus, it is possible to increase the identification range of the RFID system with an energy reduction of one bit, so that P _inuk will be ≥P _in1 .

In RFID system readers based on identifiers on SAWs, coding sequences are any code from 1 to 2 ^N , where N is the number of bits:

The reader receiving device receives coding signals distorted by noise. In the optimal receiver, the input signal is “compared” using a correlator with all kinds of pairs and the “most similar” one is selected. Due to noise, the input coding signal can be converted into one of the possible codes. This will happen when some of the units are distorted to zeros and / or some zeros are distorted to units. The most likely will be a conversion to a code that differs by one unit or zero.

The communication range in such a reading system, among other things, will depend on the probability of distortion of the received code in any neighboring one. The probability of distortion will depend on the "distance" between the code sequences. The "distance" between the code sequences is equal to the number of mismatched bits.

In the system for implementing the prototype method, code sequences are a set of consecutive binary numbers, with a minimum code distance of one. For example, a code sequence consists of 64 characters (bits). It is enough to distort one of the characters, and we get a false identification.

Thus, a minimum difference between codes of one bit increases the probability of a code recognition error and reduces the identification range.

The disadvantage of the prototype method is the high probability of a code recognition error, which is equivalent to a decrease in the range.

The objective of the proposed technical solution is to increase the maximum range of the radio-frequency identification system by encoding identifier data with selected code sequences that differ by more than one bit.

To solve the problem in a method for increasing the range of a radio frequency identification system, including using a binary code to obtain code sequences used to encode identifier data, according to the invention, the code sequences are selected to differ by at least a predetermined number of bits more than one, and the code sequences generate or by enumerating codes on a computer, or using block codes, or using a generator ora convolution codes.

The proposed method includes the use of binary code to obtain code sequences used to encode identifier data that are selected to differ by at least a predetermined number of bits (characters) more than one.

Selected code sequences are implemented on identifiers on a SAW by selective application of reflectors on a chip.

A set of code sequences can be obtained in several ways:

- sequential search of all codes for the presence of a code distance equal to that specified by a computer;

- using a block code generator with a given code distance, using, for example, the Reed-Solomon code;

- using an ultra-precise code generator.

Getting a set (array) of sequences with a code distance (the number of different bits) is more than one.

1. Direct enumeration on a given property.

The search algorithm is as follows.

Any initial code is selected; from the remaining ones, codes are selected with the number of differing bits greater than the set, for all pairs of codes from the initial and selected in the second step, codes are selected that differ from the first two by the specified number of bits. This code selection algorithm continues until all codes from a given number of 2 ^N are exhausted, where N is the number of bits. The selection procedure is repeated for all codes selected as initial. Thus, as a result of enumerating code sequences to the difference between any of the codes for a given number of bits, we obtain sets (arrays) of codes with this property.

From the obtained sets of codes, sets with the maximum number of codes are selected to provide more identifiers. This algorithm requires large computational resources, increasing with increasing N. For large values of N, its implementation is possible on a powerful computer. The advantage of the method is the ability to use it for any value of N and obtain the maximum set of codes with a given code distance.

2. Using a block code generator.

Consider the generation of codes with a given code distance for Reed-Solomon block codes. The Reed-Solomon code has the following structure:

(n, k) = (2 ^m -1.2 ^m -d _min ),

where n is the number of complete code in m-bit characters,

k is the length of the encoded code,

m is any number

d _min - the minimum value of d.

For a given value N = n · m and d _min , n, k, and m are selected. For the obtained value of nk, we select a generating polynomial of degree nk and based on it we implement an encoder based on a shift register (Fig. 8.9, Digital Communications, Bernard Sklyar).

Submitting all kinds of codes of length k · m to the encoder input, we obtain at the encoder output a set of code sequences with a selected value d _{min of the} minimum code distance. This method is convenient for generating code sequences in a regular form, but it is limited in choosing the length of the code sequence.

3. Using a convolutional code generator.

For convolutional codes, optimal generating polynomials are known that realize the maximum value of the code distance for the given code parameters (Table 8.2.1., Table 8.2.11, Prokis). From the tables we select the given value of the code distance and based on the code parameters we build a convolutional encoder (Fig. 8.2.1., Prokis CA). Submitting a complete set of codes to the encoder input, we obtain coding sequences with a given code distance at the output. This method is more optimal compared to using a block code generator, but also has a restriction on the length of code sequences.

The implementation of the proposed method can be shown by the example of constructing the optimal receiver, the scheme of which is presented in Fig. 3, where s _b is the minimum data code; s _l is the maximum data code; s _i is the current data code.

Thus, in the proposed method, the selected code sequences with a code distance of more than one are used as code sequences for encoding these identifiers, which leads to a significant decrease in the probability of code recognition errors and an increase in the identification range.

Information sources

[1] Dshkhunyan V.L. Electronic identification. Contactless electronic identifiers and smart cards / V.L.Dshkhunyan, V.F.Shangin. - M .: ACT Publishing House LLC: NT-Press Publishing House, 2004. - 695 pp., Ill.

[2] Prokis D. Digital Communications. Per. from English / Ed. D.D. Klovsky. - M .: Radio and communication. 2000 .-- 800 p .: ill.

[3] RF patent 2344437, G01S 13/00.

[4] RF patent 2253149, G01S 13/00.

Claims (1)

A method of radio frequency identification based on surface acoustic waves, including the use of a binary code to obtain code sequences used to encode identifier data, characterized in that the code sequences are selected to differ by at least a predetermined number of bits, more than one, to ensure recognition errors are reduced the identifier code of the radio frequency identification system, determining the increase in its range, while the code sequence nosti generate or by sorting on a computer codes, or using block codes or convolutional codes using the generator.

Images