RU2731207C1 - Method for increasing processing efficiency of ultra-wideband short pulse signals at a receiving side - Google Patents

Abstract

FIELD: radio equipment.

SUBSTANCE: invention relates to a method of increasing processing efficiency of ultra-wideband short pulse signals at a receiving side. Method provides for a process of weight processing of pulse sequences of information UWB signals received from a radio channel and mixed with channel noise, by continuous evaluation of pulsed energy values averaged over all pulses of the pulse-width energy in three current-shifted current time windows over the duration of their duration: in current "advanced" time window, in current "signal" time window and in current "delayed" time window.

EFFECT: enabling increase in probability-time characteristics of systems and facilities using WBI signals, due to reliable retention of synchronism state between receiving and transmitting parts of radio lines, both stationary and mobile high-speed information transmission systems, multi-user systems and other radioelectronic systems and facilities in the process of radio exchange, as well as by using a dynamic energy threshold, the introduction of which partially compensates for the negative effect of external interference factors.

1 cl, 9 dwg

Description

The invention relates to the field of radio engineering and can be used in the development of receiving devices that provide an increase in the efficiency of detecting and distinguishing information UWB signals as due to the reliable retention of the state of synchronism between the receiving and transmitting parts of radio lines, both stationary and mobile high-speed information transmission systems, multi-user systems and other radio-electronic systems and means practicing radio exchange with short-pulse ultra-wideband (UWB) signals, and through the use of a dynamic energy detection threshold.

It is known from the theory and practice of using complex signals that in order to ensure high-quality reception of information codograms (messages), it is necessary that in the process of radio exchange between correspondents (subscribers) the synchronization error of the transmitting and receiving devices of these correspondents does not exceed the limit values, beyond which a failure in the operation of the radio link is possible. ... This is especially true for radio electronic systems and means using short-pulse UWB signals, since due to the very short pulse duration of such signals and the time windows in which they are detected on the receiving side, maintaining the synchronism state is a rather difficult problem. At the same time, the use of a dynamic energy threshold, the introduction of which partially compensates for the negative influence of external interference factors, increases the robustness (noise immunity) of such systems and means.

At present, the method of threshold processing of UWB signals at the receiving side has become widespread, as the simplest to implement and allows to obtain high technical transmission rates, providing an increase in the throughput of radio channels with a sufficiently high reliability of the received information. One of such methods is described in [1]. A characteristic of this method is the formation of UWB signals using interval coding of a small number of high-energy pulses with a very large average repetition period. This approach provides, on the one hand, a low average energy density of the UWB signal in the channel, and on the other hand, an excess of the amplitude of the UWB signal pulses at the receiving point of the noise level. Processing of pulses of UWB signals on the receiving side is carried out in "signal" and "noise" time windows of short duration, which provides good protection against impulse noise. In this case, the UWB signal above the "signal" threshold in the "signal" time windows should retain its structure, ie, be detected by the presence of the above-threshold part of its pulses. In this case, maintaining the synchronism state during processing of UWB signals is ensured by increasing the steepness of the leading and trailing edges of each of the UWB signal pulses.

The choice of the comparison threshold and the selection of UWB signals pulses is carried out by processing the channel noise in the "noise" time windows. Consequently, the level of the detection threshold of UWB signal pulses in the "signal" time windows is here determined by the noise level that has exceeded the "noise" threshold in the "noise" time windows.

The disadvantages of this method include the following. First, the dependence of the level of UWB signals and the quality of synchronism during processing on the degree of distortion of the pulse shape (in particular, on the steepness of the edges) at the receiving point, which limits the use of the method by the presence of channel frequency dispersion, which greatly distorts the pulse shape. Second, the “signal” threshold is adapted to the level of external noise and interference, but only to the extent that they exceed the “noise” threshold in the “noise” time windows. Moreover, it does not depend on the energy level of the UWB signal pulses. This reduces its sensitivity to the negative effects of external interference factors.

In [2], a method is presented for processing a mixture of pulses of UWB signals with channel noise using the energy detection threshold of the UWB signal. This is justified by the fact that very often the shape of the UWB signal pulses is a priori unknown; therefore, the only sign of its presence in the channel is its energy, which can only be detected in the process of processing a mixture of a UWB signal with channel noise by comparing the received UWB signal energy with the energy threshold.

Detection of a UWB signal is mainly carried out by the presence of an energy peak exceeding a certain threshold. In this case, the detection efficiency will depend on the rules for choosing this threshold. Since, under conditions of a changing or unknown energy signal-to-noise ratio, a fixed value of the detection threshold can lead to a missing signal or to a false alarm, it was proposed in [2] to introduce an adaptation of the energy detection threshold to a change in the levels of both signal and noise by introducing, the so-called "dynamic" threshold, which allows to partially compensate for the negative influence of changing external interference factors on the reliability of the received information. It was found that the "dynamic" threshold depends only on the signal-to-noise ratio and does not depend on the signal level.

The main disadvantage of this method can be considered the absence or impossibility of performing operations that make it possible to compensate for the loss of the synchronism state during the radio exchange between subscribers.

The method described in [3] involves splitting the entire UWB signal into K time-disjoint intervals for their subsequent cepstral processing (used for signals that are a convolution of two time functions, which, after converting them into a spectrum, form pulses that do not overlap on the abscissa axis). But before it, it is necessary to stretch in time each such interval, using for this delays, for the implementation of which K delay lines are required. In this case, the maximum value of the cepstrum (signal characteristic, energy spectrum of the function) is found in each of the K cepstral arrays, the obtained maximum values of the cepstrum are divided by K-1, and the results are taken as threshold values for the corresponding cepstral arrays. By a special comparison of the values of these thresholds with each value from the corresponding cepstral arrays, a decision is made either on the presence of encoded logical ones and zeros in these intervals, from which the resulting array is created, which is a decoded codogram (information message), or on their absence, which means making a decision on the end of receiving the codogram.

The disadvantage of this method is that the algorithm for its implementation is rather complicated, time-consuming and requiring additional technical costs.

The closest in technical essence to the proposed is the method described in [4], taken as a prototype.

The prototype method includes an algorithm for accumulating in two current "signal" time channels the energy of bursts of UWB pulses carrying information zeros and ones, from the moment the synchronism state is captured in accordance with the following relationship

, (1)

, (1)

where u ₀ (t) is the reasonably chosen form of the UWB signal pulse with duration τ ₀ ; T _s - the duration of the UWB signal; T _un is the pulse repetition period of the UWB signal carrying the channel symbol corresponding to the information unit, s; T _zer is the repetition period of the UWB signal pulses carrying the channel symbol corresponding to the information zero, s; ν (t) - white Gaussian noise; N is the number of pulses in the UWB signal. In this case, the detection threshold (P) of the energy maximum, which makes it possible to obtain the required value of the probability of correct detection of the UWB signal, can be selected in a standard way, for example, by setting the value of the probability of a false alarm (false detection of a UWB signal) that provides a minimum value of the probability of missing a UWB signal. In the absence of UWB signals, this threshold can be periodically updated to take into account, as a first approximation, the effect of channel noise on the processing of UWB signals.

Maintaining the synchronism state is carried out during the processing of these information UWB signals in accordance with the algorithm for accumulating the energy of the basic (reference or synchronizing) pulses of UWB signals in the "leading", "central" and "lagging" time channels, which is similar to algorithm (1), but describes work of a temporary discriminator

in the "advanced" time channel,

(2)

(2)

in the "central" time channel,

in the "lagging" time channel.

Here, T _bas is the basic repetition period of the synchronizing (or reference) pulses of UWB signals, relative to which the pulses characterizing logical units (T _un = T _bas + τ _adv ) and zeros (T _zer = T _bas - τ _ret ) are located at given time positions , where τ _adv is the time shift, which ensures that the code units are followed before the corresponding reference pulse, and τ _ret is the time shift, which ensures that the logical zeros are followed after the corresponding reference pulses; τ _del is the time shift between the "leading", "central" and "lagging" time channels that form the time discriminator.

After digitizing the current energies (1) and (2) accumulated over the duration of the i-th UWB signal, we obtain the values E _{un, i} , E _{zer, i} , E _{adv, i} , E _{cen, i} , E _{ret, i} , above-threshold parts which are the arguments for the function displaying the required time offset of the "signal" time channels (1) relative to the moment of capturing the synchronism state

(3)

(3)

Since the direction of displacement of all five time channels in time is the same, then when the energy of the sync pulses of the current UWB signal accumulated according to (3) passes from the “central” time channel to the “advanced” or “lagging” time channel of the time discriminator, the value and sign of this time displacement δt _i = f (ΔE _{adv, i} , ΔE _{cen, i} , ΔE _{ret, i} ) with the subsequent introduction of this displacement in (1) and (2) in order to compensate it to increase the energy of information UWB signals accumulated according to (3) E _{un, i} , E _{zer, i} , which increases the probability of their correct detection.

From the analysis of the prototype method, it follows that to generate UWB signals carrying information zeros and ones, it uses position-pulse modulation (PIM, in the Latin version of PPM - Pulse-Position Modulation), when the reference (synchronizing) pulses of the UWB signal follow equal distances from each other on the time axis, and logical zero or logical one are located to the left and / or right of the reference pulse at selected distances.

The prototype method has the following disadvantages. First, under the conditions of the influence of random external interference factors, the values of the above-threshold energies accumulated according to (3) ΔE _{adv, i} , ΔE _{cen, i} , ΔE _{ret, i} from signal to signal with a high probability can vary to a large extent. But the magnitude and sign of the time shift δt _i are estimated using precisely these energies accumulated over the duration of the UWB signal, therefore, with a high probability, the time shift δt _i in some time intervals will be estimated with a large error, which will entail a loss of reliability accepted in the "signal" time information channels. Secondly, this method does not provide for the formation of a dynamic detection threshold of UWB signals, the introduction of which could partially compensate for the negative influence of random external interference factors by stabilizing the value of the above-threshold energy of detected UWB signals.

The task of the proposed method is to increase the reliability, as well as to reduce the influence of negative external interference factors on the information received in the "signal" time channels.

To solve this problem, in the method of increasing the efficiency of processing ultra-wideband short-pulse signals on the receiving side, using for this the organization of time channels shifted relative to each other in time by a given amount, according to the invention, the minimum number of time channels is selected, equal to three, which allows to ensure the declared reception qualities by weight processing of the energy of a mixture of pulses of current direct and inverse ultra-wideband signals with channel noise, accumulated over the duration of the current "leading", "signal" and "lagging" time windows, formed in the corresponding time channels, including:

- estimation of the value of the accumulated impulse energies;

- estimation of the value of the above-threshold parts of the impulse energy readings;

- digitization of the current accumulated impulse energies, which are further referred to as impulse energy readings;

- fixing the time positions of the above-threshold impulse energy readings;

- reduction of non-zero values of above-threshold impulse energy readings to units;

- estimation of the sums accumulated on the duration of direct or inverse ultra-wideband signals of single digital samples;

- comparison of the time positions of the above-threshold pulse energy samples obtained by accumulating energy over the duration of the current "signal" time windows of the corresponding time channel with the time positions of single samples characterizing the time positions of the pulses of direct and inverse ultra-wideband signals and recorded in the corresponding registers;

at the same time, if the sums of digital single samples accumulated on the duration of the current "leading" time windows is less than half of the number of pulses in an ultra-wideband signal, but more than the same amount, but accumulated on the duration of the current "lagging" time windows, then it is assumed that the first mentioned sum is zero ; if the second said sum of digital unit samples is less than half of the number of pulses in the ultra-wideband signal, but more than the same sum mentioned first, it is assumed that the second said sum is zero; if the sums of digital single samples accumulated on the duration of the current "signal" time windows of the corresponding time channel do not exceed the value of the selected digital threshold, then the current ultra-wideband signal will be skipped; if the indicated digital sums exceed the selected digital threshold, but the time positions of more than half of the corresponding pulse energy samples do not coincide with the time positions of the unit samples stored in the corresponding register, then the current ultra-wideband signal will be skipped again; if, in addition to the above-mentioned digital sums, the digital threshold will be exceeded, but the time positions of more than half of the corresponding impulse energy samples coincide with the time positions of the unit samples stored in the corresponding register, then in this case the current ultra-wideband signal will be received and defined as an information unit or information zero, depending on whether it is a direct ultra-wideband signal or inverse; simultaneously:

- calculated, reduced to the duration of time windows, the modules of time position differences between the above-threshold pulse energy samples of the "advanced" and "signal" time windows, as well as between the above-threshold pulse energy samples of the "lagging" and "signal" time windows;

- fixed non-zero reduced differences of any value;

- their number is estimated;

- the reduced differences of a unit value are recorded;

- their number is estimated;

- the sign of the difference between the number of non-zero above-threshold readings obtained by processing in "lagging" time windows and the number of non-zero above-threshold readings obtained by processing in "advanced" time windows of the corresponding time channels when processing direct or inverse ultra-wideband signals is estimated;

- the above-threshold parts of the current pulses of direct or inverse ultra-wideband signals located in the "leading" and "signal" time windows or in the "signal" and "lagging" time windows are calculated, reduced to the pulse duration, depending on the direction of the shift of the time windows relative to the moment of entering synchronism with subsequent assessment of their average values;

- an assessment of the degree of harmful influence of channel noise on the reliability of the obtained average values and compensation of this influence is carried out;

- on the basis of the results of the calculations performed, the time shift of the time windows of the corresponding time channels is estimated relative to the moment when the synchronism state is captured with the subsequent correction of this shift;

- the value of the current dynamic threshold is estimated with its subsequent use in the processing of ultra-wideband signals coming from the radio channel.

и

для обнаружения в них импульсов соответственно прямого и инверсного информационных СШП сигналов на приёмной стороне, где моменты начала формирования этих текущих временных окон также задаются с помощью «разреженного» кода Неймана-Хоффмана: чёрные прямоугольники соответствуют текущим временным позициям «сигнальных» временных окон для обнаружения в них импульсов прямого СШП сигнала, а серые прямоугольники - текущим временным позициям «сигнальных» временных окон для обнаружения в них импульсов инверсного СШП сигнала.In the claimed invention, pulse-code modulation (CPM) is used, in which UWB signals carrying information units and zeros are formed as follows. A certain code is selected (here the 24-element Neumann-Hoffman code is the most balanced), the pauses between the pulses of this code are proportional to the numbers of the pseudo-random number sequence, which has the same number of elements as the selected code. For a UWB signal carrying an information unit (hereinafter referred to as a direct UWB signal), positions corresponding to units of the Neumann-Hoffman code are selected as the time positions of the pulses, and for a UWB signal carrying an information zero (hereinafter inverse UWB signal), positions are selected as time positions corresponding to zeros of the Neumann-Hoffman code. FIG. 1a) shows the "sparse" Neumann-Hoffman code NH (t), on the basis of which direct and inverse UWB signals are formed, where the black filled circles correspond to the time positions of the pulses of the direct UWB signal, and empty squares - to the time positions of the pulses of the inverse UWB signal. FIG. 1b) the current time positions of the "signal" time windows are presented as a function

and

to detect in them pulses, respectively, of direct and inverse information UWB signals on the receiving side, where the moments of the beginning of the formation of these current time windows are also set using the "sparse" Neumann-Hoffman code: black rectangles correspond to the current time positions of the "signal" time windows for detection in them are pulses of the direct UWB signal, and the gray rectangles are the current time positions of the "signal" time windows for detecting in them pulses of an inverse UWB signal.

The explanatory graphics are presented in the following figures.

и

; фиг. 2а) – текущие временные позиции этих временных окон в своих временных каналах для обработки импульсов прямого и инверсного СШП сигналов; фиг. 2б) – укрупнённый фрагмент, содержащий три соответствующих временных окна для обработки прямого СШП сигнала; фиг. 3а) и фиг. 3в) – представлены величины энергетических отсчётов, накапливаемых импульсных (оконных) энергий смеси импульсов прямого (фиг. 3а) и инверсного (фиг. 3в) СШП сигналов с канальными шумами на своих временных позициях; фиг. 3б) и фиг. 3г) – величина энергетических отсчётов, характеризующих надпороговые части накопленных в «опережающих» временных окнах импульсных энергий; фиг. 4 (пояснение к фиг. 3) – укрупнённый масштаб ситуации до накопления текущей энергии прямого (фиг. 4а) и инверсного (фиг. 4б) СШП сигналов в «сигнальных» временных окнах своих временных каналов; фиг. 5 динамика накопления единичных цифровых отсчётов, с меньшей (фиг. 5а) и большей (фиг. 5б) физической энергетикой; фиг. 6 - фиг. 8 характеризуют те же особенности представленного способа обработки СШП сигналов на приёмной стороне, что и на фиг. 3 – фиг. 5, но в случае использования текущего значения динамического энергетического порога П_д,i и текущей оценки временного смещения δt_i; на фиг. 9 представлена укрупнённая блок-схема устройства для реализации предлагаемого способа.FIG. 1a) - "sparse" Neumann-Hoffman code NH (t); fig. 1b) - current time positions of "signal" time windows as a function

and

; fig. 2a) - current time positions of these time windows in their time channels for processing pulses of direct and inverse UWB signals; fig. 2b) - enlarged fragment containing three corresponding time windows for processing the direct UWB signal; fig. 3a) and Fig. 3c) - shows the values of energy counts, accumulated pulse (window) energies of a mixture of pulses of direct (Fig. 3a) and inverse (Fig. 3c) UWB signals with channel noise at their time positions; fig. 3b) and Fig. 3d) - the value of the energy counts characterizing the above-threshold parts of the impulse energies accumulated in the "advanced" time windows; fig. 4 (explanation to Fig. 3) is an enlarged scale of the situation before the accumulation of the current energy of the direct (Fig. 4a) and inverse (Fig. 4b) UWB signals in the "signal" time windows of their time channels; fig. 5 dynamics of the accumulation of single digital readings, with less (Fig. 5a) and more (Fig. 5b) physical energy; fig. 6 to FIG. 8 characterize the same features of the presented method for processing UWB signals on the receiving side as in FIG. 3 to FIG. 5, but in the case of using the current value of the dynamic energy threshold P _{d, i} and the current estimate of the time shift δt _i ; in fig. 9 shows an enlarged block diagram of a device for implementing the proposed method.

The proposed method is as follows. The time discriminator processes the incoming sequences of mixtures of pulses and channel noise directly from the direct and inverse information UWB signals in three time channels, using algorithms for the accumulation of pulse energies in the corresponding current time windows, designated in their time channels as "leading", "signal" and "lagging" ... In this case, the "signal" time window is actually the receiving time window. FIG. 2a), conventionally in the form of rectangles of dark gray, black and light gray colors, respectively, the current time positions of these time windows are presented in their time channels for processing pulses of direct and inverse UWB signals. FIG. 2b), for greater clarity, an enlarged fragment is presented containing only three corresponding time windows for processing the direct UWB signal. Pre-processing of UWB signals is as follows.

In the corresponding time channel for the duration of the current "signal" time windows:

- the values of the accumulated impulse (window) energies of the mixture of impulses of UWB signals with channel noise or only noise are estimated

(4)

(4)

и

- the above-threshold parts of the digitized energies are calculated

and

(4^/)

(4 ^/ )

- the current impulse energies (4) and (4 ^/ ) are digitized;

, а их ненулевые значения приводятся к единицам- time positions of values are fixed (4 ^/ )

, and their nonzero values are reduced to units

(4^//)

(4 ^// )

- the sum of accumulated current single digital readings is estimated, which is the equivalent of the maximum of the classical energy autocorrelation function in digital representation

(4^///)

(4 ^/// )

, то при наличии в канале СШП сигнала он будет пропущен, так как не будет преодолён цифровой порог П_ц принятия решения о типе СШП сигнала. При этом осуществляется переход к оконной обработке следующего СШП сигнала информационной кодограммы. - if a

, then if there is a UWB signal in the channel, it will be skipped, since the digital threshold _{Pc of} making a decision on the type of UWB signal will not be exceeded. In this case, the transition to window processing of the next UWB signal of the information codogram is carried out.

, а временные позиции

или

энергетических отсчётов, полученных в текущих «сигнальных» временных окнах, совпадают с временными позициями единичных отсчётов, записанных в регистр, характеризующих прямой и инверсный СШП сигналы и представленных на фиг. 1а), то осуществляются следующие операции:If a

and temporary positions

or

the energy samples obtained in the current "signal" time windows coincide with the time positions of the single samples recorded in the register characterizing the direct and inverse UWB signals and presented in Fig. 1a), then the following operations are performed:

- based on the results (4 ^/// ), a decision is made about which of the information symbols was accepted, one or zero;

- in the corresponding time channel, on the duration of the "advanced" time windows, the values of the accumulated impulse (window) energies of the mixture of UWB signals with channel noise or only noise are estimated

(5)

(five)

и

- the above-threshold parts of the digitized energies are calculated

and

(5^/)

(5 ^/ )

- the current impulse energies (5) and (5 ^/ ) are digitized;

, а их ненулевые значения приводятся к единицам- time positions of values are fixed (5 ^/ )

, and their nonzero values are reduced to units

(5^//)

(5 ^// )

- the sum of the accumulated current single digital readings is estimated

(5^///)

(5 ^/// )

при

, то принимается

или

в силу относительной ненадёжности оценок; - if a

at

, then it is accepted

or

due to the relative unreliability of estimates;

- in the corresponding time channel on the duration of the "lagging" time windows, the values of the accumulated impulse (window) energies of the mixture of UWB signals with channel noise or only noise are estimated

(6)

(6)

и

- the above-threshold parts of the digitized energies are calculated

and

(6^/)

(6 ^/ )

- the current impulse energies (6) and (6 ^/ ) are digitized;

, а их ненулевые значения приводятся к единицам- time positions of values are fixed (6 ^/ )

, and their nonzero values are reduced to units

(6^//)

(6 ^// )

- the sum of the accumulated current single digital readings is estimated

(6^///)

(6 ^/// )

при

, то принимается

или

также в силу относительной ненадёжности оценок.- if a

at

, then it is accepted

or

also due to the relative unreliability of estimates.

– величина текущего расстояния между импульсами инверсного СШП сигнала;

– числа из псевдослучайной числовой последовательности, определяющие величины

для инверсного СШП сигнала;

для прямого СШП сигнала и

для инверсного СШП сигнала;

для прямого СШП сигнала и

для инверсного СШП сигнала – определяют степень различия уровней шумов в «шумовых» временных окнах; ξ – величина, от которой зависит верхний предел интегрирования для каждого члена квадратов сумм из (4)-(6), то есть для членов, содержащих только импульс СШП сигнала ξ = τ₀, а для членов, содержащих либо чисто шумовые члены, либо содержащих произведения шумов с импульсами ξ = 2τ₀;

- энергии импульсов прямых и инверсных СШП сигналов, накапливаемые в «опережающих», «сигнальных» и «запаздывающих» временных окнах, соответственно, в своих временных каналах;

- взаимные энергии импульсов и канальных шумов прямых и инверсных СШП сигналов, накапливаемые в этих же временных окнах;

- шумовые энергии, накапливаемые одновременно в этих же текущих временных окнах; П₀ – значение опорного энергетического порога, определённое в процессе захвата состояния синхронизма; round(x) – округление величины х до ближайшего целого числа. Все оцифрованные импульсные энергии будем называть соответствующими энергетическими отсчётами.Here t_k - the moments of the beginning of the formation of the current time windows for processing a mixture of pulses of direct and inverse UWB signals with channel noises, which are the most important of the temporal parameters appearing in the proposed method; T_k = b_kτ₀ - the value of the current distance between the pulses of the direct UWB signal; b_k - numbers from a pseudo-random numerical sequence that determine the values of T_k for direct UWB signal;

- the value of the current distance between the pulses of the inverse UWB signal;

- numbers from a pseudo-random numerical sequence that determine the values

for inverse UWB signal;

for direct UWB signal and

for inverse UWB signal;

for direct UWB signal and

for an inverse UWB signal - determine the degree of difference in noise levels in "noise" time windows; ξ is the quantity on which the upper limit of integration depends for each term of the squares of the sums from (4) - (6), that is, for terms containing only the UWB signal pulse ξ = τ₀, and for terms containing either purely noise terms or containing products of noise with momenta ξ = 2τ₀;

- energy of impulses of direct and inverse UWB signals, accumulated in "advanced", "signal" and "lagging" time windows, respectively, in their time channels;

- mutual energies of pulses and channel noise of direct and inverse UWB signals, accumulated in the same time windows;

- noise energies accumulated simultaneously in the same current time windows; P₀- the value of the reference energy threshold, determined in the process of capturing the state of synchronism; round (x) - Rounds x to the nearest integer. All digitized impulse energies will be called corresponding energy readings.

Simultaneously:

- calculated, reduced to the duration of time windows, the modules of time position differences between impulse energy samples of the "leading" and "signal" time windows, as well as between impulse energy samples of the "lagging" and "signal" time windows

- fixed non-zero reduced differences of any value

- their number is estimated

- reduced differences of a unit value are recorded

- their number is estimated

) СШП сигналов- the sign of the difference between the number of non-zero above-threshold samples obtained by processing in "lagging" time windows and the number of non-zero above-threshold samples obtained by processing in "advanced" time windows of forward (zn) or inverse (

) UWB signals

- taking into account the results (5 ^/// ) and (6 ^/// ), calculate the reduced durations of the parts of the current pulses of the direct or inverse UWB signals located in the "leading" and "signal" time windows for the first case, and in the "signal" and "Lagging" time windows for the second case

(7)

(7)

(8)

(8)

- averaging (smoothing) of values (7) and (8) is performed

(9)

(nine)

(10)

(ten)

- an assessment of the degree of harmful influence of channel noise on the reliability of results (9) and (10) is carried out, at which the differences

are not equal to zero, therefore it is necessary to compensate for this effect, for which the contribution (weight) of this harmful effect on each of the quantities (9), (10) is determined

(11)

(eleven)

(12)

(12)

- based on the estimate of weights (11), (12), the values (9), (10) are corrected as follows

(13)

(13)

(14)

(fourteen)

then, if operations (11) - (14) were performed without errors, the check should give

- if the current direct UWB signal is present in the radio channel, the estimation of the time offset to correct the "drift" of the time windows of the corresponding time channels is carried out as follows

(15)

(15)

(16)

(sixteen)

(17)

(17)

- in the case of the presence of the current inverse UWB signal in the radio channel, the estimation of the time offset for correcting the "drift" of the time windows of the corresponding time channels is carried out similarly to (15) - (17)

(18)

(18)

Simultaneously with the above actions, operations are performed that lead to an estimate of the value of the current dynamic threshold P _{d, i} , which will be used in the process of receiving the remaining UWB signals of the information codogram instead of the reference P ₀ :

и

- since the direct and inverse UWB signals cannot appear in the radio channel at the same time, the current time windows of mutually inverse UWB signals are used when processing their pulse sequences as "noise" time windows in which the corresponding impulse noise energies are accumulated to estimate the current values

and

in this case, the obvious relations are used

(19)

(nineteen)

and taking into account (19), with sufficient accuracy for practice, we can say that

that is, the measured (estimated) values are close enough to the true values of the current values;

- the evaluation of the signal-to-noise ratio values averaged over the duration of the current UWB signals is carried out

(20)

(20)

- the value of the current dynamic threshold P _{d, i} is estimated, based on the fact that the value of the probability of correct detection of the current UWB signal is a function that depends on the current value of the energy threshold for detecting pulses in the current time windows, on the total number of pulses in it and on the number of its undetected pulses at a given value of the quantity (20), while for any allowable number of undetected pulses the probability of correct detection of the current UWB signal has its own value of the energy threshold, which maximizes this probability, then the current dynamic threshold P _{d, i} can be defined as a value proportional to the average the value of energy thresholds that maximize the probability of correct detection of the current UWB signal when the number of undetected pulses in it changes from one to a certain maximum allowable number for a given current value of the value (20);

- the reference threshold is replaced by the current value of the dynamic threshold, which minimizes the negative influence of variable external interference factors, and the obtained estimates (17) and (18) are introduced into the upper and lower limits of integration in relations (4), (5) and (6), which determine the results of the accumulation of impulse energies in the current "leading", "signal" and "lagging" time windows, correcting the moments of the beginning of their formation in their time channels, while the value of the above-threshold energy of the pulses of information UWB signals detected in the current "signal" time window is optimized, and in the current "leading" and "lagging" time windows - minimized

(21)

(21)

FIG. 3a) and Fig. 3c) as an illustration, the values of energy samples, accumulated impulse (window) energies of a mixture of pulses of direct (Fig.3a) and inverse (Fig.3c) UWB signals with channel noise at their time positions described by relations (4) - (6) ... There is also given the reference energy threshold P ₀ , which is depicted by a bold horizontal dashed line in dark gray color. Black filled circles indicate energy counts accumulated in "signal" time windows, black empty squares - energy counts accumulated in "advanced" time windows, black filled squares - energy counts accumulated in "lagging" time windows. 3b) and FIG. 3d) shows the above-threshold parts of these energy readings, (4 ^/ ) - (6 ^/ ). Here, dark gray filled rhombuses represent the time positions of copies of UWB signal pulses stored for comparison in the form of corresponding samples. The image shown in FIG. 4 serves as an explanation to FIG. 3, as it illustrates the enlarged scale of the situation before the accumulation of current energy (curves of dark gray color)

(22)

(22)

direct (Fig. 4a) and inverse (Fig. 4b) UWB signals in the "signal" time windows (black rectangles) of their time channels. From an analysis of FIG. 4, it follows that the time windows begin to lag behind the moment of entering synchronism, therefore, the current energy of the mixture of UWB signals with channel noise begins to be redistributed between the “leading” (dark gray rectangles) and “signal” time windows. It is for this reason that FIG. 3b) and Fig. 3d) the value of the energy counts characterizing the above-threshold parts of the impulse energies accumulated in the "advanced" time windows is so great, and the spread of the values of the energy counts characterizing the above-threshold parts of the impulse energies accumulated in the "signal" time windows is so significant.

FIG. 5 as confirmation of the analysis of FIG. 3 and FIG. 4 shows the dynamics of the accumulation of single digital samples, the maxima of which are determined by the relations (4 ^/// ) - (6 ^/// )

(23)

(23)

Moreover, in FIG. 5b) black and gray rectangles conventionally represent the current time positions of the "signal" time windows for processing direct (black) and inverse (dark gray) UWB signals in their time channels. Analysis of FIG. 5a) and 5b) shows that the outwardly digital form of estimating the maximum of the autocorrelation function from (4 ^/// ) is, as it were, insensitive to the shift of time windows, since the one shown in Fig. 5a) is indistinguishable from that shown in FIG. 5 B). However, these images cannot be viewed separately from FIG. 3b) and Fig. 3d), since from the analysis of the material presented there it follows that the single readings dynamically accumulated according to (23) and presented in Fig. 5a) are characterized by much lower physical energetics than similar readings shown in Fig. 5 B). This means that the real probability of correct detection of UWB signals, the pulses of which are processed in the current "signal" time windows is much higher than the same, but conditional (illustrative) probability, when the pulse processing is carried out in the current "advanced" time windows.

The illustrative material shown in FIG. 6 to FIG. 8, characterizes the same features of the presented method for processing UWB signals on the receiving side as in FIG. 3 to FIG. 5, but in the case of using the current value of the dynamic energy threshold P _{d, i} and the current estimate of the time shift δt _i . In this case, the relationships for describing the dynamics of the accumulation of single digital readings are similar to relationships (23), but taking into account (21).

The analysis shown in FIG. 6 to FIG. 8 and comparing it with that shown in FIG. 3 to FIG. 5 shows that the use of the current estimate of the time offset for the intended purpose sets with high accuracy the pulses of the forward and inverse UWB signals at the center of the "signal" time windows of their time channels, and the introduction of the current dynamic energy threshold takes into account the increased energy of the samples in the "signal" time windows of the corresponding time channels. This entails the following:

- increasing the pulse energy accumulated on the duration of the "signal" time windows to the maximum values;

- reduction of the spread of the values of energy readings, characterizing the above-threshold parts of the pulse energies accumulated in the "signal" time windows, to the minimum values;

- zeroing of possible residual noise energy samples in "leading" and "lagging" time windows of their time channels.

From the presented analysis, it follows that the implementation of the method proposed in the invention for processing information UWB signals on the receiving side will allow continuously and with a minimum error in the process of radio exchange between correspondents to maintain the synchronism state, maximize the probability of correct detection of UWB signals, minimize the probability of false alarms (false detections of UWB signals) and will allow keeping the value of the probability of passing the UWB signal within the required boundaries. Therefore, the declared qualities will be ensured.

An enlarged block diagram of a device for implementing the proposed method is shown in Fig. 9, where the following designations are introduced:

1.1 - 1.3 - the first, second and third time channels (VK);

2 - analog-to-digital converter (ADC);

3 - processing and control unit (BOU);

4 - synchronization unit (BS).

The device contains three time channels 1.1 - 1.3, an analog-to-digital converter 2, a processing and control unit 3 and a synchronization unit 4. In this case, the first inputs of VK 1.1 - 1.3 are combined and are the input of the device. The second inputs of VK 1.1 - 1.3 are connected to the corresponding outputs of the BOU 3, the output of which is connected to the input of the synchronization unit 4, the first, second and third outputs of which are connected to the third inputs of VK 1.1 - 1.3, respectively. The VK 1.1 - 1.3 outputs are connected by buses with the corresponding inputs of the ADC 2, the output of which is connected by a bus to the first input of the ACU 3, the second input of which is connected to the fourth output of the BS 4. In addition, the input / output of the ACU is the input / output of the device.

The device works as follows. When the command to switch to the mode of receiving UWB signals is received at the input / output of the BOU 3, the signal to start operation will be sent to the input of the BS 4 from the output of the BOU 3, the clock pulses necessary for control will begin to arrive at the second input of the BOU 3 from the fourth output of the BS 4 at a given pace by the state of VK 1.1 - VK 1.3, the second inputs of which from the corresponding outputs of the BOU 3 will receive signals that turn them on, at the same time, the third inputs of VK 1.1 - VK 1.3 from the corresponding outputs of BS 4 will begin to receive synchronizing clock pulses at an appropriate pace to control time positions "Leading", "signal" and "lagging" time windows corresponding to these time channels. The first inputs of VK 1.1 - VK 1.3 will begin to receive either channel noise or a mixture of UWB sync pulses with channel noise. In this case, from the outputs of VK 1.1 - VK 1.3, the energies (4) - (6) and (4 ^/ ) - (6 ^/ ), accumulated in the current time windows VK 1.1 - VK 1.3, will come to the corresponding inputs of ADC 2, and from the output of ADC 2 these accumulated energies will be supplied to the first input of the BOU 3 sequentially in accordance with the order of work of the time windows VK 1.1 VK - VK 1.3. In BOU 3, the digitized energies (4 ^/ ) - (6 ^/ ) in the form of the corresponding energy readings will be used to estimate the time displacement (17), (18) of the corresponding time windows VK 1.1 - VK 1.3 relative to the moment of entering synchronism, and the digitized energies ( 4) - (6) - to estimate the value of the current dynamic threshold. The obtained estimates at the end of the duration of the current UWB signal are fed to the second and third inputs of VK 1.1 - VK 1.3 from the corresponding outputs of the BOU 3 and BS 4, correcting the time positions of the "leading", "signal" and "lagging" time windows corresponding to their time channels, simultaneously adapting the energy threshold to the current interference environment. When the next UWB signal arrives at the input of the device, the described procedure is repeated. This process ends when the signal of the end of the radio exchange between the correspondents arrives at the input / output of the BOU 3.

The implementation of the device implementing the proposed method does not cause difficulties, since the functional units included in the units of the device are well known, are widely represented in domestic and foreign patents, and are also described in the technical literature. All devices described in the enlarged block diagram are given, for example, in [4] and [5].

Sources of information

1. Ageikin, V.I. On the issue of using the technology of ultra-wideband signals in the interests of creating promising communications, reconnaissance and electronic warfare of the tactical control link. Ageikin, L.M. Kaplyarchuk, A.P. Stepanov // Electronic warfare in the Armed Forces of the Russian Federation, part 1 - 2017, p. 40-44.

2. Popov, A.S. Detection of impulse signals in wireless subscriber access networks / A.S. Popov, V.A. Kovtun, V.A. Salamov // Scientific Almanac. Technical sciences - 2016, No. 2-2 (16), p. 385-388.

3. RF patent 2416162. Asynchronous method of extracting encoded information transmitted to the consumer using ultra-wideband pulses. IPC H04B 7/00. Zhbanov I.L., Silaev N.V., Mitrofanov D.G., Senkov M.A., Zhbanova V.L., Vasilchenko O.V., Gavrilov A.D. Application No. 2009146425/09 dated 14.12.2009. Publ. 20.06.2010

4. Kornienko A.V. Algorithms for the synthesis and processing of short-pulse ultra-wideband signals in radio systems for transmitting information, taking into account interfering factors. / Abstract of a dissertation for the degree of candidate of technical sciences. - Ryazan. - 2008 .-- S. 17.

5. RF patent 2315424. Communication system with high speed of information transmission by ultra-wideband signals. IPC Н04B 1/69, H04L 5/26. Bondarenko V.V., Kyshtymov G.A., Bondarenko V.V., Kyshtymov S.G. Application No. 2006119887/09 dated 06.06.2006. Publ. 20.01.2008

Claims (19)

A method for increasing the efficiency of processing ultra-wideband short-pulse signals on the receiving side, using for this the organization of time channels shifted relative to each other in time by a given value, characterized in that the minimum number of time channels is selected, equal to three, which allows to ensure the declared reception qualities by weighting energy a mixture of pulses of current direct and inverse ultra-wideband signals with channel noise, accumulated over the duration of the current "leading", "signal" and "lagging" time windows, formed in the corresponding time channels, including: - estimation of the value of the accumulated impulse energies; - estimation of the value of the above-threshold parts of the impulse energy readings; - digitization of the current accumulated impulse energies, which are further referred to as impulse energy readings; - fixing the time positions of the above-threshold impulse energy readings; - reduction of non-zero values of above-threshold impulse energy readings to units; - estimation of the sums accumulated on the duration of direct or inverse ultra-wideband signals of single digital samples; - comparison of the time positions of the above-threshold pulse energy samples obtained by accumulating energy over the duration of the current "signal" time windows of the corresponding time channel with the time positions of single samples characterizing the time positions of the pulses of direct and inverse ultra-wideband signals and recorded in the corresponding registers; in this case, if the sums of digital single samples accumulated on the duration of the current "leading" time windows is less than half of the number of pulses in the ultra-wideband signal, but more than the same amount, but accumulated on the duration of the current "lagging" time windows, then it is assumed that the first mentioned sum is zero; if the second said sum of digital unit samples is less than half of the number of pulses in the ultra-wideband signal, but more than the same sum mentioned first, it is assumed that the second said sum is zero; if the sums of digital single samples accumulated on the duration of the current "signal" time windows of the corresponding time channel do not exceed the value of the selected digital threshold, then the current ultra-wideband signal will be skipped; if the indicated digital sums exceed the selected digital threshold, but the time positions of more than half of the corresponding impulse energy samples do not coincide with the time positions of the unit samples stored in the corresponding register, then the current ultra-wideband signal will be skipped again; if, in addition to the above-mentioned digital sums, the digital threshold is exceeded, but the time positions of more than half of the corresponding impulse energy samples coincide with the time positions of the unit samples stored in the corresponding register, then in this case the current ultra-wideband signal will be received and defined as an information unit, or information zero, depending on whether it is a direct ultra-wideband signal or inverse; simultaneously: - calculated, reduced to the duration of time windows, the modules of time position differences between the above-threshold pulse energy samples of the "advanced" and "signal" time windows, as well as between the above-threshold pulse energy samples of the "lagging" and "signal" time windows; - fixed non-zero reduced differences of any value; - their number is estimated; - the reduced differences of a unit value are recorded; - their number is estimated; - the sign of the difference between the number of non-zero above-threshold readings obtained by processing in "lagging" time windows and the number of non-zero above-threshold readings obtained by processing in "advanced" time windows of the corresponding time channels when processing direct or inverse ultra-wideband signals is estimated; - the above-threshold parts of the current pulses of direct or inverse ultra-wideband signals located in the "leading" and "signal" time windows or in the "signal" and "lagging" time windows are calculated, reduced to the pulse duration, depending on the direction of the shift of the time windows relative to the moment of entering synchronism with subsequent assessment of their average values; - an assessment of the degree of harmful influence of channel noise on the reliability of the obtained average values and compensation of this influence is carried out; - on the basis of the results of the calculations performed, the time shift of the time windows of the corresponding time channels is estimated relative to the moment when the synchronism state is captured with the subsequent correction of this shift; - the value of the current dynamic threshold is estimated with its subsequent use in the processing of ultra-wideband signals coming from the radio channel.

Images