RU2760567C1 - Method for generating and detecting a sync pulse of a noise-like signal that does not depend on the correlation properties of the sequences modulating the phase of the signal - Google Patents

Abstract

FIELD: radio engineering.

SUBSTANCE: invention relates to radio engineering and can be used as a detector of phase-manipulated noise-like signals. In the method for generating and detecting a synchro pulse of a noise-like signal on the transmitting side, the sequence (DPse) by which the phase of the carrier frequency of the synchro pulse is modulated is formed by bitwise multiplication of one of the Walsh functions (FWk), with one of the sequences of the complete code (DPn_se). On the receiving side, the DPh sequence is calculated by bitwise multiplication of the DPse and DPn_se sequences and the correlation functions of this sequence with all Walsh functions. The decision to detect a synchro pulse is made only if the maximum of the correlation function (CF) is achieved with the FWk and the sum of the modules of the calculated CF, except for the CF of the sequence of DPh with the FWk, is less than Δ, which is determined by the accuracy of the representation 0.

EFFECT: reduction in false detections.

1 cl, 1 dwg, 2 tbl

Description

The invention relates to the field of communication and can be used in systems in which it is necessary to use noise-like phase-shift keyed signals.

In such systems, information is transmitted using service and information elements. The service element of the noise-like signal (sync pulse) is used to detect it. With the help of information elements, the necessary information is transmitted. Modulation of the carrier phase of the service and information elements is performed by binary sequences.

Phase-shift keyed signal sync pulse detection is performed using the classic Detector, which consists of a Matched Filter, an Adaptive Threshold Determiner and a Solver.

The operation of such a Detector is based on the calculation of the correlation function (CF) of the currently observed signal with the sequence that modulated the phase of the sync pulse. An adaptive threshold finder calculates the level of the currently observed signal. When the CF value of the adaptive threshold value is exceeded, the Solver fixes the detection of the signal.

The disadvantage of this Detector is the "false detection" associated with:

- poor correlation properties of the binary sequences used (ACF of the modulating sync pulse sequence and CCF of this sequence with modulating sequences of information elements);

- an increase in the side peaks of the modulating sequences due to external interference;

- the method for calculating the adaptive threshold;

- the presence of internal noise at the input of the receiving device.

It is possible to minimize the number of "false detections" in such a Detector only if such sequences are selected for which the ACF and CCF have small values of the side peaks. It is still possible to select reference sequences for modulation of the sync pulse that have side peaks of the ACF less than the specified value. But to select for the selected reference sync pulse sequence a set of sequences for modulating signal information elements, for which the maximum CCF peak is equal to a given value, is a very difficult, practically impossible computational task (for a base of a signal element n = 256, it is necessary to calculate 2 ²⁵⁶ ≈ 10 ⁷⁷ CCF and choose from them those for which the maximum peak does not exceed the specified value). In addition, the selection of binary sequences with good correlation properties significantly reduces their number.

The choice of one or another method for calculating the adaptive threshold depends on the nature of external interference, and therefore is unpredictable. For the same sync pulse, due to the randomness of the interference characteristics, the adaptive threshold can be underestimated, which can lead to "false detection", or overestimated, which can lead to signal skipping.

A decrease in the intensity of the internal noise of the receiving device leads to a deterioration in the sensitivity of the communication system.

On "false detection" processing of "false information items" is started. In the best case, the "detection falsity" can be determined at the stage of processing the first "false information element". In the worst case, a "detection falsity" can be determined by processing the last "false information item". This reduces the bandwidth of the communication system.

In the patent RU 2505934 C1 (H04L27 / 233, 01/27/2014) it is proposed to combat "false detections" by dynamically changing the threshold value in accordance with certain signal parameters at the output of the matched filter. To do this, the average Cp is calculated and the peak Yn value of the signal at the output of the matched filter is measured, after which the value of the specified threshold is calculated using the formula: P = v * (Yn-Cp) + Cp, where v is a constant proportional to the relative value of the side peaks of the ACF reference sequence.

The disadvantage of this method of dealing with "false detections" is that its implementation requires the calculation of Cp and Yn, and then the value of P, which requires certain hardware and time. In addition, each DP requires its own value of v, which should be selected based on the magnitude of the side peaks of the ACF.

The closest in technical essence to the claimed method is a method for combating "false detections", given in patent RU 2608769 C1. In this patent, the phase of the sync pulse is proposed to be modulated with the sequence:

where FU _k is the Walsh function that plays the role of the feature "detection by the main peak", DPn_si is the sequence with which the FU _k is "noisy", and the symbol "*" means bitwise multiplication, such that:

1 * 1 = 1, -1 * -1 = 1, -1 * 1 = -1, 1 * -1 = -1.

The detection method proposed in this patent is a classical sync pulse detector, supplemented by a block for calculating the "detection by the main peak" feature. The “detection” feature will be generated only in the case of the simultaneous occurrence of two events: the CF value has exceeded the adaptive threshold (“excess” signal) and the “main peak detection” feature has been developed.

Such a Detector eliminates "false detections" associated with poor correlation properties of the used binary sequences, as well as with "false detections" associated with the presence of an Adaptive Threshold Device (low threshold).

The disadvantage of such a Detector, as well as those discussed above, is the skipping of the sync pulse. This is due to the presence in the considered Detectors of the Adaptive threshold device (the case of overstating the threshold).

In the claimed detection method, it is proposed to form a sync pulse in the same way as in patent RU 2608769 C1, and to detect a sync pulse using a Detector, the operation of which is similar to the operation of a Decoder that processes signal information elements. The only difference is that the input of the Decoder, which processes the information elements of the signal, always receives one of the Walsh functions, while the input of the Decoder, which plays the role of the Detector, can receive either the Walsh function or any of the sequences of the complete code that is not a Walsh function.

The functional diagram of the claimed Detector is shown in Fig. one.

.The receiver receives and digitizes the input information that goes to the input of the Detector. The detector works in a step-by-step manner. In each clock cycle, it processes n signal samples represented by

...

*ДПн_си (Декодер обработки информационных элементов вычисляет ФУ_у=

*ДПн_иэ).The detector calculates DPx =

* DPn_si (The decoder for processing information elements calculates the FU _y =

* DPn_ie).

The DPx processing unit calculates the CF of the DPx sequence with all sequences of the FU _j . (The processing unit of the Information Element Decoder calculates the CF of the DPu sequence with all sequences of the FU _j ). Determines the maximum value of CF (DPx, FU _j ) (The processing unit of the Information Element Decoder calculates the maximum CF (DPu, FU _j )).

.If the maximum CF is not reached with FU _k , then the Detector proceeds to processing the next portion

...

If the maximum value of the CF of the DPx sequence with Walsh functions is achieved with the FU _k and the sum is:

.then the Detector generates the "detection" sign. If (2) is not met, then the Detector proceeds to processing the next portion

...

To substantiate such a criterion for the development of the "detection" feature, let us consider some properties of binary sequences from which this criterion follows.

Property 1.

It is known that FUs are orthogonal.

This means that the CF of any sequence of a complete code with Walsh functions reaches its maximum value only in the only case when this sequence is a CF _k . The maximum CF value in this case will be equal to n, and the CF values of this sequence of the complete code with FU ₀ , FU ₁ , ..., FU _k-1 , FU _{k + 1} , ..., FU _n will be equal to 0.

The CFs of any sequences of a complete code that are not a Walsh function with Walsh functions will have values less than n (they are not orthogonal to the FU).

For example, for the base n = 8, Table 1 shows the CF values of the sequences DP185, DP193, DP196, DP195, DP85, DP96, DP105, with Walsh functions. FU _{6 is} taken as FU _k .

Table 1 DPx CF values of the DPx sequence with FU ΣKF FU ₀ FU ₁ FU ₂ FU ₃ FU ₄ FU ₅ FU ₆ FU ₇ DP185 2 2 -2 6 2 2 -2 -2 eight DP193 -2 -2 2 2 2 2 6 -2 eight DP196 -2 -2 6 -2 2 2 2 2 eight DP195 0 0 0 0 0 0 eight 0 eight DP85 0 -eight 0 0 0 0 0 0 -eight DP96 -4 0 0 -4 4 0 0 -4 -eight DP105 0 0 0 0 0 0 0 -eight -eight

It can be seen from Table 1 that the maximum CF value is achieved at DP195 (this is FU ₆ !) And is equal to 8, and the CF values of the DP195 sequence with FU ₀ , FU ₁ , ..., FU ₅ , FU ₇ are equal to 0. All other CF sequences with Walsh functions less than 8.

Property 2.

The sum of the CF of any sequence of a complete code with Walsh functions is equal to either "-n" or "n".

Table 1 shows that ΣKF (DP185, FU _j ) = ΣKF (DP193, FU _j ) = ΣKF (DP196, FU _j ) = ΣKF (DP195, FU _j ) = 8, and ΣKF (DP85, FU _j ) = = ΣKF (DP96, FU _j ) = ΣKF (DP105, FU _j ) = - 8.

Taking into account the properties described above, let us explain the operation of the Detector with signals, the base of which is n = 8. For such signals, Table 1 could have, for example, the form presented in Table 2.

table 2 DPx CF values of the DPx sequence with FU ΣKF FU ₀ FU ₁ FU ₂ FU ₃ FU ₄ FU ₅ FU ₆ FU ₇ DP185 22.3 22.1 -22.2 65.9 21.9 22.1 -21.9 -22.0 88.2 DP193 -22.1 -21.9 22.1 22.2 21.9 21.9 66.1 -22.2 88 DP196 -21.9 -22.1 66.2 -22.2 21.1 22.2 22.1 21.9 88.3 DP195 3 * 10 ^-5 2 * 10 ^-5 4 * 10 ^-5 1 * 10 ^-5 5 * 10 ^-5 1 * 10 ^-5 87.9 2 * 10 ^-5 87.900 DP85 1 * 10 ^-5 -88.2 5 * 10 ^-5 3 * 10 ^-5 2 * 10 ^-5 5 * 10 ^-5 1 * 10 ^-4 4 * 10 ^-5 -88.199 DP96 -43.9 3 * 10 ^-5 4 * 10 ^-5 -44.2 44.1 2 * 10 ^-5 8 * 10 ^-5 -43.9 -87.899 DP105 8 * 10 ^-5 6 * 10 ^-5 2 * 10 ^-5 5 * 10 ^-5 7 * 10 ^-5 9 * 10 ^-5 98 * 10 ^-5 -87.8 -87.799

Note.

For real signals, there will be no CF = 0 values. This is due to digitization accuracy, internal noise and signal strength. For example, "zero" can be represented by a value of 0.00001 (or much less). Then, in the case of DPx = FUk, all CF (DPx, FUj), except for CF (DPx, FUk), will not be equal to 0, but will have values of the order of 0.000001. In the worst case, if all these CF are equal to 0.00001, then their sum will be equal to Δ = 0.00001 * (n-1). For n = 8, 16,…, 256, the value of Δ will be equal to 0.00007, 0.00015,…, 0.00255, respectively.

.It can be seen from Table 2 that the appearance at the input of the DPx processing unit of signals, the phase of which is modulated by the DP185, DP196, and DP96 sequences, the maximum CF values are not achieved with FU ₆ . In these cases, the Detector will proceed to processing the next portion.

...

For signals whose phase is modulated by the sequences DP193, DP195, DP85 and DP105, the maximum CF value is achieved with FU ₆ . In these cases, the Detector calculates the sum of all modules of the obtained values of CF (DPx, FU _j ), except for CF (DPx, FU _k ).

=22.1,

=21.9,

=22.1,

=22.2,

=21.9,

=21.9,

=22.2. Сумма этих значений будет равна 154.3, что больше 0.00007, следовательно, условие выработки признака «обнаружение» не выполняется.For DPx = DP193 values

= 22.1,

= 21.9,

= 22.1,

= 22.2,

= 21.9,

= 21.9,

= 22.2. The sum of these values will be equal to 154.3, which is more than 0.00007, therefore, the condition for generating the "detection" feature is not met.

For the same reason, the "detection" sign will not be generated when the DPh sequence DP85 and DP105 appears at the input of the DPx processing unit.

=0.000003,

=0.000002,

=0.000004,

=0.000001,

=0.000005,

=0.000001,

=0.000002. Сумма этих значений равна 0.000018, что меньше 0.00007, следовательно, будет выработан признак «обнаружение».For DPx = DP195 values

= 0.000003,

= 0.000002,

= 0.000004,

= 0.000001,

= 0.000005,

= 0.000001,

= 0.000002. The sum of these values is 0.000018, which is less than 0.00007, therefore, the "detection" feature will be generated.

If the “detection” feature appears at the input of the Information Element Detector, the “detection” sign will not occur if there is no DPSI sequence among the modulating sequences of information elements.

Similarly, the declared Detector will work with signal bases 2 ¹⁶ , 2 ³² ,…, 2 ²⁵⁶ ,….

The claimed method for generating and detecting a sync pulse of a phase-shift keyed signal allows:

- significantly improve the throughput of the communication system due to the absence of "false detections" generated by the poor correlation properties of the modulating sequences;

- to improve the operation of the communication system when it is exposed to various kinds of interference (no influence of external interference on the calculation of the threshold, and, therefore, there will be no gaps in the sync pulse);

- allows to use for modulation of signal elements all sequences of a complete code;

- significantly simplify the hardware (the hardware of the Detector is similar to the hardware of the Information Element Decoder).

Claims (1)

A method for generating and detecting a sync pulse of a noise-like signal, in which, on the transmitting side, the formation of a sequence (DPSI), with the help of which the phase of the carrier frequency of the sync pulse is modulated, is carried out by bitwise multiplication of one of the Walsh functions (PCF), with one of the sequences of the complete code (DPSN_SI), and on the receiving side, by bitwise multiplication of the DPSi and Dpn_si sequences, the Dpx sequence and the correlation functions of this sequence with all Walsh functions are calculated, and the decision to detect a sync pulse is made only if the maximum of the correlation function (CF) is reached with the PFK and at the same time the sum of the modules of the calculated CFs, except for the CF of the sequence DPx with FUk, less than Δ, which is determined by the representation accuracy 0.

Images