# Device for signals processing

FIELD: information technologies.

SUBSTANCE: device comprises serially connected frequency filter, digitiser and unit of reduction to perfect instrument (RPI), intended for interpolation of counts supplied to its inlet, detection of weight of basic final duration of signals in inlet signal on the basis of interpolated counts decomposition into Fourier series by orthogonalised reactions of frequency filter into basic signals and for formation of outlet signal as a superposition of basic signals with account of their weight in inlet signal, besides versions of device include connection of noise suppression unit or serialy connected unit of signal growth speed assessment and normalisation unit between digitiser and RPI unit.

EFFECT: improved resolution and sensitivity to elements of signal, increased efficiency and simplification of device for signals processing.

4 cl, 12 dwg

The proposed device relates to the field of signal processing by solving the inverse problem to improve their compliance with the input device signals and, primarily, to the field of signal processing to suppress intersymbol interference and a corresponding increase resolution of their elements, increase the speed of data transfer at the same physical bandwidth of the device.

Known [1] the device for processing signals containing serially connected frequency filter (BSF) and the optimal regenerating the filter, allowing you to Gaussian noise minimum RMS error recovery. The disadvantage of these devices is the lack of resolution of the signal elements.

Known [2] the device for processing signals containing consistently connected the black sea fleet, discretization and the convolver and provides approximately twice higher resolution signal elements. These devices have the drawback consisting in a high level oscillations (parasitic oscillations, "tails", side lobes) outside the subject of the resolution elements of the input signal and low contrast elements and groups of elements of small length.

Known [3] the device for processing signals containing consistently connected the black sea fleet, discretize the PRS and the interpolator, in which the representation of signals in the basis spheroidal wave functions and the corresponding higher concentration of energy in the low frequencies provide some reduction in the level of spurious oscillations and increase the resolution. These devices have the drawback consisting in not sufficiently reduced level of spurious oscillations in low contrast maloprodajnih elements signal and insufficient resolution of the signal elements.

The common feature of these devices is to maximize the resolution of the signal elements within the available physical bandwidth that is only slightly reduces inter-symbol interference (ISI) and what caused these basic weaknesses. More promising in terms of increasing the resolution of the signal elements of the following devices, in which actions in a limited frequency band is more broadband the result.

Known [4] the device for processing signals containing consistently connected the black sea fleet, discretization and block the restoration of the signal Monte Carlo, providing not limited by the bandwidth of the BSF steepness of the gradient and the distance resolution. The disadvantage of such devices is the low probability of matching the restored signal is input, since the principle of their action is based on the Integra is Inoi information about the signal (energy) not the input device and the input for the block restoration.

The known device [5] for processing signals containing consistently connected the black sea fleet, discretization and block reduction device to quasisovereign intended for interpolation of samples of the signal, identifying differences and the formation of a reduced description of the signal sequence differences with characteristics close to the characteristics of fluctuations in the input signal. The disadvantage of such devices is the lack of sensitivity to low-contrast elements of the signal, low speed and high complexity, due to the principle of the synthesis signal on the basis of detection of changes and their correction by comparing the results of the reactions of the black sea fleet on the input and the synthesized signals. Namely, some differences, especially small contrast, can be missed due to their masking the oscillations, that is due to the intrinsic noise of the device; a low speed and the high complexity associated with the need to compute the response of the black sea fleet on the synthesized signal and the need for its comparison with the reaction of the black sea fleet on the input signal.

From this lack of free device [6] for processing signals containing consistently connected the black sea fleet, discretization and block reduction device to be the resultant, producing a substitute passed through the device narrowband signal elements of the original broadband signal elements, acting on the input of the system. It directly and drastically reduces inter-symbol interference and accordingly increases the resolution of the signal elements, the capacity of the device. In other words, in the signal processing unit reduction device to perfect the device becomes a virtual band, which is wider than the original physical bandwidth of the device.

However, these devices have the disadvantage, consisting not of a high enough resolution signal elements, a lack of sensitivity to low-contrast elements of the signal in the low speed and high complexity, due to the principle, which is heuristic (non-deterministic) nature of a substitute passed through the device narrowband signal elements of the original broadband signal elements, which is an iterative type of control procedure is complete according to the criterion of achieving processed by the signal pre-defined or dynamically determined characteristics. Namely, the elements of the lesser contrast may be missed due to the lower accuracy of signal representation in iterative calculations, that is because of the noise the device; low speed and high complexity due to the heuristic, iterative nature of the calculations and the need for funds, supervising the completion of the synthesis of the recovered signal.

The present invention aims to reduce the self-noise of the device by eliminating the heuristic iterative calculations, as well as the exception of funds, supervising the completion of the synthesis of the recovered signal.

The solution to this problem provides:

- increase the resolution of the signal elements, including low-contrast;

- increased sensitivity to low-contrast elements signal;

- improve performance;

- reducing the complexity of the device.

(In the first variant of the proposed device) in the device for processing signals containing consistently connected the black sea fleet, the entrance of which is an input device, and discretization, additionally, we introduce the block reduction to the perfect device (block RBC)used for interpolation arriving at its input from discriminator counts, determine (using the estimates of the coefficients of the decomposition of aggregate interpolirovanii times in a generalized Fourier series in samples orthogonalizing measured or stored reactions frequency filter on the basis of finite duration, but the options input device signal) factors weight basis of finite duration elements (signals) in the input signal and generation timing of the output device signal samples evaluation input for device signal by superposition of basis signals taking into account their weight coefficients in the input signal (cumulative summation of the products of samples of the basic elements of the input signal and the weight coefficients of the basis elements in the input signal), and the input unit RRU is connected to the output of discriminator, and the output is an output device. This allows discretization to get the samples passed through the BSF signal in the unit of RBC to interpolate from discriminator counts, the interpolated samples using samples orthogonalizing reactions frequency filter on the basis of finite duration, the elements of the input device signal to determine the weight coefficients of the basis elements in the input signal and by the cumulative summation of the counts of the basic elements of the input signal, multiplied by the weight coefficients of the basis elements in the input signal, to generate samples of the output signal in the form of counts of the evaluation of the input device signal, thereby to project a tolerance of performing perfect broadband (with suppressed inter-symbol interference) restore signal up to the maximum permitted and with an extremely small amplitude of its elements.

The input device representable continuous PREOBRAZOVANIYa signal f(x) can be described acceptable (with a design tolerance) his assessment (x) as a superposition

where p_{i}. - weight coefficients;

α_{i}(x) - reference signals (items) end lengths (durations). Accordingly, the output response of the black sea fleet (BSF together with sensitive elements of discretization) on the signal f(x) and assessment(x) is

where ⊗ is the convolution operation;

ψ(x) is the impulse response of the black sea fleet (pulse function, impulse response function of the scattering point).

From (3) it follows that the weights p_{i}are the expansion coefficients of the signalin a generalized Fourier series on the basis signalsif the latter is orthogonal.

Therefore, the proposed device for achieving the goals of the inverse problem (the problem of determining the estimate of the input signal in response to the BSF) is solved by the following steps:

a) orthogonalization [7]

(in the General case needed) reactionsthe underlying signalswhere

{·} is the designation of many elements;

〈·〉 - operation orthogonalization;

b) determination of the coefficients

decomposition reactions

(a(x)b(x))=

A - scope signal a(x) and b(x);

C) determining evaluationthe input signal in the form of (1) at

where s_{i}- weight coefficients (relative levels) signalsspecific their views. Under other equal conditions the preference is given to methods of orthogonalization, providing a value of s_{i}=1.

Joint consideration of (1) and (3) allows to interpret the RBC as the principle of substitution narrowband signalsbroadband reference signals a(x).

On the restoration of the input device for processing signals, the method described is asimptoticheski (with the increase in the detail of the underlying signals a(x)) perfect. Regarding recovery estimates (1) of the input signal, this device is perfect regardless of its degree of approximation to the input signal.

The limiting resolution "plot" elements of the signal f(x) in the assessment(x) is limited to the adopted set of basic signals α_{i}(x), the step of sampling the recovered signal is La and noises,
reducing the accuracy of determination of the Fourier coefficients of k_{i}. The above design tolerance on the choice of evaluation(x) of the input device signal takes into account the error in the determination of the coefficients of p_{i}.

A set of basic signals in the simplest case may be, for example, from a combination of rectangular pulses with different quantitative characteristics; a more complex set may consist of, for example, from the Haar functions [4] or from a combination of, for example, a rectangular pulse, a triangular pulse, pulses of arbitrary shape under various quantitative characteristics.

Interpolation preceding the determination of the Fourier coefficients is, for example, in accordance with the interpolation formula

theorem [4] about the samples. Here

n=1, 2, ..., N - number of signal samples(x) at the output of the interpolator;

N is the number of signal samples(x) at the output of the interpolator;

m - number of signal samples(x) at the input of the interpolator;

M - number of signal samples(x) at the input of the interpolator;

Δ - sampling signal(x);

δ=Δ/q - step interpolation;

q≥1 - coefficient interpolation;

h(x) - Interpol is the dominant core of the form h(x)=sinc(πx/Δ); sinc(x)=sin(x)/x.

When the orientation of the device only on the value of q=1, the interpolation function can be absent.

The second version of the proposed device is that the device for processing signals containing consistently connected the black sea fleet, the entrance of which is an input device, and discretization, inputs connected in series block noise reduction designed to reduce noise through the normalization of the incoming input samples of the impulse elements with characteristics not corresponding to the characteristics of possible output responses of the black sea fleet, and the unit of RBC, intended for interpolation arriving at its input from discriminator counts, determine (using the estimates of the coefficients of the decomposition of aggregate interpolirovanii of samples in the Fourier series for samples orthogonalizing measured or stored reactions frequency filter basic finite duration elements of the input device signal) factors weight basis elements in the input signal and the generation timing of the output device signal by superposition of basis signals taking into account their weight coefficients in the input signal (in the form of counts of the evaluation input device signal by the cumulative summation produced the response times of the basic elements of the input signal and the weight coefficients of the basis elements in the input signal), the block noise reduction is connected with the output of discriminator, and the output unit of RBC is the output device. This allows the block noise reduction to reduce noise level by normalizing the incoming output discriminator counts of pulse elements with characteristics not corresponding to the characteristics of possible output responses of the black sea fleet, and in the RBC unit to perform the above for the first variant of the device of the image forming timing of the output device signal. Reduction block noise reduction of the noise level further enhances achievable in the device resolution of the input signal for him.

The third version of the proposed device is that the device for processing signals containing consistently connected the black sea fleet, the entrance of which is an input device, and discretization, inputs connected in series interpolator, an input connected to the output of discriminator, the determinant (using estimates of the coefficients of the decomposition of aggregate interpolirovanii of samples in the Fourier series for samples orthogonalizing measured or stored reactions frequency filter on the basis of finite duration, the elements of the input device signal) weights basic finite duration elements (signal is) in the input device signal and the driver counts the output device signal samples evaluation input for device signal by superposition of basis signals with regard to their coefficients weight in the input signal (by cumulative summation of the products of samples of the basic elements of the input signal and the weight coefficients of the basis elements in the input signal)and the output of the shaper counts is the output device. This allows discretization to get the samples passed through the black sea fleet of the input signal, the interpolator to generate a detailed representation of the processed signal up to a scale of the smallest subject to resolution elements in the determinant of the weights of the underlying signals in the input signal to determine the weight coefficients of the basis elements in the input device signal (using the estimates of the coefficients of the decomposition of aggregate interpolirovanii times in a generalized Fourier series in samples orthogonalizing measured or stored reactions frequency filter on the basis of finite duration, the elements of the input device signal) to calculate the weight coefficients of the basis elements in the input signal, and the driver counts the output device signal by superposition of basis signals with respect coefficients of their weight in the input signal (by cumulative summation of counts of the basic elements of the input signal, multiplied by the weight coefficients of the basis elements in the input signal), to calculate from the couple the output device signal, thus with the design tolerance forming reaction of the perfect device for input.

The fourth version of the proposed device is that the device for processing signals containing consistently connected the black sea fleet, the entrance of which is an input device, and discretization, inputs connected in series, the node estimates the growth rate of a signal, an input connected to the output of discriminator, the Normalizer, the interpolator, the determinant (using estimates of the coefficients of the decomposition of aggregate interpolirovanii of samples in the Fourier series for samples orthogonalizing measured or stored reactions frequency filter on the basis of finite duration, the elements of the input device signal) weights basic finite duration elements (signals) in the input device signal and the driver counts the output for device signal by superposition of basis signals taking into account their weight coefficients in the input signal (in the form of counts of the evaluation input device signal by the cumulative summation of the products of samples of the basic elements of the input signal and the weight coefficients of the basis elements in the input signal)and the output of the shaper counts is the output device. This allows discretize the Torah to receive the samples passed through the black sea fleet of the input signal, in the site evaluation of growth rate and node normalization to verify compliance of the pulse elements possible output responses of the black sea fleet, to normalize unacceptably distorted pulse elements and thereby provide for more efficient operation of the interpolator, the determinant of the weights of the basis elements (signals) in the input signal and driver samples the output signal that described above for the third variant of the device bearing the weight of the basis elements in the input signal, and calculates the timing of the output device signal, thereby forming (with a design tolerance) reaction of a perfect device for the input signal under conditions of low noise level.

In figure 1, figure 2, figure 3, figure 4 shows the block diagram of the proposed device for processing signals according to paragraphs 1, 2, 3, 4 claims.

Figure 5 presents an example block diagram of discretization.

Figure 6 presents an example block diagram of the RBC unit.

Figure 7 presents an example block diagram of the block noise reduction (block PP).

On Fig presents an example block diagram of the interpolator.

Figure 9 presents an example block diagram of the determinant of the coefficients of the weight of the basis elements in the input signal (the determinant of weights).

Figure 10 presents an example block diagram of the imaging unit samples the output signal (forsirovannogo signal).

Figure 11 presents an example block diagram of a node estimates the rate of growth (site assessment).

On Fig presents an example block diagram of the Normalizer.

The first version of the device contains the BSF 1, discretization 2, block 3 RBC.

The second variant of the device contains the BSF 1, discretization 2, block 3 RBC, block 4 PP.

The third variant of the device contains the BSF 1, discretization 2, the interpolator 5, the determinant of the 6 scales, the driver 7 of the output signal.

The fourth variant of the device contains the BSF 1, discretization 2, the interpolator 5, the determinant of the 6 scales, the imaging unit 7 output node 8 of the evaluation, the Normalizer 9.

CF represents, for example (in the simplest case, when processing the temporalx input signals), resistive-capacitive low pass filter.

Discretization 2 contains, for example, the pulse generator 10, the imaging unit 11 grid frequency, analog-to-digital Converter (ADC) 12.

Unit 3 RBC contains, for example, the interpolator 5, the determinant of the 6 scales, the driver 7 of the output signal (when the interpolation factor is equal to the interpolator unit 5 performs the identical transformation, and may be absent).

Unit 4 PP contains, for example, node 8, the Normalizer 9.

The interpolator 5 contains, for example, random access memory (RAM) 13, a driver 14 argument, constantly the storage device (ROM) 15, cumulative multiplier 16.

Keys 6 scales contains, for example, a ROM 17, a ROM 18, the transmitter 19 of the coefficients of the weight.

Shaper 7 output signal contains, for example, the ROM 20, the multiplier 21, the cumulative calculator 22 samples the output signal, the RAM 23.

Node 8 assessment contains, for example, a register 24, myCitadel 25.

The Normalizer 9 includes, for example, the corrector 26 counts, item 27 of the comparison, the switch 28.

ADC 12 can be performed, for example, in the form of chips AD9461 [8].

ROM 15, a ROM 17, a ROM 18, ROM 20 can be made on the basis of, for example, chip MRR [9].

The RAM 13, the RAM 23 can be made on the basis of, for example, chip RU [9].

The input devices:

29 - information.

Output device:

30 - information output.

The inputs and outputs of the component parts, which inputs and outputs an integral part of a higher level, have the numbers of inputs and outputs integral part of a higher level.

Other inputs (discriminator 2, block 3 RBC, block 4 noise suppression, the determinant of the 6 parameters of the elements of the input signal shaper 7 output signal Normalizer 9):

31 - signal input of discriminator 2,

32 - input unit 3 RRU,

33 - unit 4 PP,

34 - the sign of the determinant of the 6 scales,

35 to the input of the shaper 7 output

36 - input Normalizer 9.

The OS is the real outputs (BSF 1, discriminator 2, block 4 PP, interpolator 5, the determinant of the 6 scales, unit 8 assessment):

37 output BSF 1,

38 - out of discriminator 2,

39 output unit 4 PP,

40 is the output of the interpolator 5,

41 - exit identifier 6 weights

42 output node 8 of the assessment.

The internal input bus - sources constants defined by the connection lines to a logical "0" or "1":

43 code maximum growth rate,

44 code of safety factor on growth rate,

Figure 1-12 thin lines presents a single connection to send individual signals, thick lines - group communication groups of signals.

To simplify the description of the following is the processing of one-dimensional signal.

In the initial state, all variants of the device on their information input 29 no signal (valid signal is equal to zero), the device handles a null signal, forming a zero signal in all informational sections and at its output 30, i.e. the device is in the dynamic state of readiness for signal processing.

The first variant of the device for signal processing works as follows. With the arrival of the signal at the input 29 of the device of the BSF 1 converts it according to their impulse response. Output BSF 1 signal with the limited spectrum is fed to the input of discriminator 2, to the which subjects it to the discretization step and the quantization for example, a uniform binary 16-bit code. Output discriminator 2 binary codes counts (ADU) of the signal fed to the input of block 3 of RBC, which interpolates q-factor of the received timing signal, the interpolated samples and the samples orthogonalizing reactions BSF 1 on the basic elements determines the weight coefficients of the basis elements in the input device signal, and these factors generates a timing output for the device signal, which is the evaluation of the input device signal, by the cumulative summation of the counts of the basis elements, multiplied by these coefficients, and thereby performs a perfect recovery of the signal up to the maximum permitted and with an extremely small amplitude of its elements. From the output of block 3 RBC formed thus the signal at the output 30 of the device.

The second variant of the device for signal processing works as follows.

With the arrival of the signal at the input 29 of the device of the BSF 1 and discretization 2 above for the first variant of the device of the image formed on the exit codes of the samples of the signal received by the input unit 4 suppression. Unit 4 noise reduction reduces the noise level by modifying the incoming output discriminator 2 counts of pulse elements with the features, not relevant to the characteristics of possible output responses of the BSF 1, and outputs the received samples in block 3 of RBC, which is described above for the first variant of the device by way perform perfect recovery of the signal and outputs it to the output device 30.

The third variant of the device for signal processing works as follows.

With the arrival of the signal at the input 29 of the device of the BSF 1 and discretization 2 above for the first variant of the device of the image formed on the exit codes of the samples of the signal that arrives at the input of the interpolator 5. The interpolator 5 performs the interpolation of these samples q-factor, i.e. generates a detailed representation of the processed signal up to a scale of the smallest subject to permission of elements and returns the interpolated samples to the determinant 6. Keys 6 on the interpolated samples calculates weight coefficients of the basis elements in the input signal using the determined identifier 6 expansion coefficients of the interpolated samples in a generalized Fourier series in orthogonalizing reactions frequency filter on the basic elements and displays them in the imaging unit 7 counts the output signal on these factors generates a timing output for the device signal, which is the evaluation of the input to give the tion signal, by the cumulative summation of the counts of the basis elements, multiplied by the weight coefficients. From the output of the imaging unit 7 counts the output signal arrives at the output 30 of the device.

The fourth variant of the device is as follows.

With the arrival of the signal at the input 29 of the device of the BSF 1 and discretization 2 above for the first variant of the device of the image formed on the exit codes of the samples of the signal that arrives at the input node 8 assessment of growth rate. Node 8 determines the growth rate of the pulse signal elements and displays its value in the Normalizer 9, which determines whether this speed is possible in the output responses of the BSF 1 and leads to normal unacceptably distorted impulse items, leaving the rest unchanged. Normalized so the samples arrive in the interpolator 5, which together with the identifier in the 6 parameters of the signal elements and the imaging unit 7 described above for the third variant of the device of the image forming timing of the output device signal. From the output of the imaging unit 7 counts the output signal arrives at the output 30 of the device.

Discretization 2 works as follows.

The generator 10 produces pulses of constant frequency, coming from its output to the input of the shaper 11 grid frequency. The imaging unit 11 generates from the pulse generator 10 is required for device operation clock pulses, coming into output bus 38 and engraved ADC 12. ADC 12 on the leading edge of each clock pulse received at its engraved, samples (reference) value of a signal from input 31 of discriminator 2 at its signal input, and converts the value into a linear binary, for example, 16-bit code received in the form of data from the ADC output 12 V output bus 38 discriminator 2. Bus 38 clock pulses fed to the input of block 3 RBC (interpolator 5, the determinant of the 6 scales of the underlying signals, the driver 7 of the output signal) and block 4 suppression (node 8 assessment of growth rate, the Normalizer 9). The necessary clock pulses transmitted by these units and all of their component parts, and because of this are present in all the information tires devices at all levels (in order not to overload the drawings, figures 1-12 broadcast clock pulses are not reflected).

Unit 3 RBC comprising as an example of the interpolator 5, the determinant of the 6 scales and the driver 7, the latter for these nodes in the third variant of the device.

Unit 4 noise suppression, comprising as an example of node 8 assessment of growth rate and Normalizer 9, the latter for these nodes in the fourth variant of the device.

The interpolator 5 operates in accordance with the expression 7) as follows.

With 32 samples passed through the BSF 1 signal produced by discretization 2, are written in the RAM 13, and then cyclically read from it (one cycle for each interpolated reference) and arrive at (informational) the cumulative input of multiplier 16. Thus from the input bus 32 to the first and second inputs of the former (14 argument come heartbeats-increments corresponding(mΔ) and the prediction samples(nδ). These pulse shaper 14 generates codes n-mq arguments counts kernel h((n-mq)δ)at the input of the ROM 15 as an address previously stored in these counts kernel computed with step interpolation. Corresponding to these addresses the samples h((n-mq)δ) of the interpolating kernel from the ROM 15 are received at the second input of the cumulative multiplier 16, which for each of the N interpolated samples produces the summation of M works(mΔ)h((n-mq)δ) with record (to clock pulses at the third entrance of the cumulative multiplier 16 by line in the composition of output bus RAM 13) each of the accumulated amount in the storage device in the cumulative multiplier 16 as the next interpolated reference frame. Output cumulative multiplier 16 interpolated samples are given (so the new impulses on the line as part of the bus exit 40) to the output 40 of the interpolator 5.

Keys 6 weights is as follows.

Input 34 of the determinant of the 6 scales of the basic elements ADU(nδ) response to the input device signal is fed to the input of the transmitter 19 of the weight coefficients. Simultaneously to the second input of the transmitter 19 output from the ROM 18 and the third input of the transmitter 19 output from the ROM 17 are timing codes (times) pre-stored in the ROM samples, respectively orthogonalizing reactionscalculated with a step of interpolation, and the values of s_{i}/(,accordingly, the expression (4, 5, 6), for example, for s_{i}=1. Along the lines comprising the bus 34 to the inputs of a ROM 17, a ROM 18 receives the signal mode of operation of these ROMs that specifies the read mode. Issuance of samples from ROM 17, a ROM 18 is made in accordance with the codes of the addresses, which consist of numbers i reaction and the sample number of this reaction and are formed in the ROM 17, a ROM 18 for incremental clock pulses of the rooms i reactions and numbers of samples with the set of codes of the addresses in the "zero" state of the reset pulses at the input of these ROMs. The sequence of receipt of these codes on the first, second and third inputs of the calculator 19 calculates it values ((nδ),and ozena the coefficients p_{
i}weight of basic signals in the form ofCalculated estimates of coefficients of weight by clock pulses from transmitter 19 are issued at the output 41 of the determinant of the 6 scales.

The imaging unit 7 counts the output signal is as follows (in ROM 20 pre-stored samples α_{i}(nδ) basis elements, specified in increments of interpolation).

Output 41 of the determinant 6 weights the coefficients p_{i}is fed to the input 35 of the imaging unit 7 counts the output signal and then input to the multiplier 21. The input of the ROM 20 lines in the composition of the input bus 35 signal is reset (set to zero) the addresses of the cells and incremental (plus 1) signals for the formation of the rooms i basic signal α_{i}(nδ) and the number n of reference of this signal. Together, these rooms are the cell addresses of the ROM 20. The samples α_{i}(nδ) baseline signals corresponding to these addresses, the output of the ROM 20 sequentially receives the second input of the multiplier 21. From the second output of the ROM 20 to the input of the cumulative calculator 22 receives the codes of rooms times the basic signals. The multiplier 21 calculates the timing works p_{i}α_{i}(nδ) and outputs them to the second input of cumulative transmitter 22. At the third entrance of the cumulative transmitter 22 to the lines in the composition of the input bus 35 receives a reset signal (set in the zero state the addresses of the operands,
incremental signals for address generation n output samples, the signals of the cumulative summation signal (sign) the issuance of the transmitter 22 samples the output signal. Cumulative transmitter 22 generates an address i+n cumulative modifiable reference to the content of my memory at this address adds the product of p_{i}α_{i}(nδ)received at its second input of the multiplier 22. After completion of the cumulative formation thus (in accordance with (1, 6)) each sample of the output signal of this reference in support of his address from the output of the transmitter 22 is issued to the input of the RAM 23 and is written in it. The reset signal address operand, the sign of the issuance of the RAM 23 samples of the output signal, the increment signal to the address generation of the n output samples received at a second input of the RAM 23 on the lines in the composition of the input bus 35, the RAM 23 counts the output signal is issued at the output 30 of the imaging unit 7, which is the output device.

Node 8 evaluation of the rate of growth is as follows.

Output discriminator 2 current samples of the signal fed to the input of 33 knots 8 assessment of growth rate and further information on the input of the register 24, to the input of reducing vicites 25 and to the output bus 42. Clock coming from bus 33 on the state clock inputs of the register 24, the samples of the signal recorded in autotragic 24 and with its output fed to the input of wichitaeagle of vicites 25 and to the output bus 42. (Previous) samples of the signal in the register 24 are entered first clock pulse (coming on the first engraved) on the leading edge, and the rest (coming in second engraved) - trailing edge. Thus, myCitadel 25 calculates the difference between the current and previous samples and outputs the obtained modules of differences, which is the current values of the rate of growth of the signal and the sign bit of the difference at the output node 42 8 assessment of growth rate; however, for the first count as prior to he used.

The Normalizer 9 operates as follows.

The values of the growth rate of the signal coming from the output node 42 8 assessment of growth rate to the input 36 of the Normalizer 9, is fed to the input element 27 comparison. With the same input 36 to the input of switch 28 receives the current samples of the signal. To the second input element 27 comparison and to the input of the offset 26 with internal thread 43 unit 9 normalization comes the code of the maximum possible growth rate of the signal due to the filter BSF 1. The result of the comparison of the rate of growth of the signal with the highest possible comes from the output element 27 comparison to the second input of switch 28. On the second and third inputs of the offset 26 from input 36 receives respectively the code from the previous reference frame and the sign of the rate of growth. Corrector 26 times, calculates the product of the poppy is Kalnay growth rate and the safety factor (smaller units), code with the internal thread 44 of the Normalizer 9 comes to the fourth input corrector 26, and considering the sign of the rate of growth adds it to the previous reference. Thus obtained corrected current reference comes from the output of the corrector 26 to the third input of the switch 28. The switch 28 when the value of the control signal from the output element 27 comparison, the corresponding allowable growth rate, produces at the output 39 of the Normalizer 9 reference input 36 unchanged, otherwise the adjusted reference value corresponding to the allowable growth rate of a signal at its third input, and thereby normalizes the failed samples.

The performance of each option device is provided at the adequacy stored in the ROM 15, a ROM 17, a ROM 18, ROM 20 data and constants to internal buses 43, 44.

The essence of the invention does not change when the redistribution of functions between components of the device changes its structure or functions that do not alter the principles of the solution of the inverse problem and noise suppression, for example, by including in its composition additional tools to streamline the presentation of the processed data, for orthogonalization reactions of the underlying signal, while refining processing modes (for example, in the real and the unrealistic time).

The proposed device for signal processing applicable to linear systems with known impulse response and the need for solving the inverse problem. The greatest effect of the proposed device for signal processing network in its application, for example, in systems, problem implemented by weight and power consumption of optical-electronic systems, including television), in systems with expensive resources for collective use (communication systems, especially broadband).

A factor of increasing resolution and data rates without expanding the physical bandwidth can reach values of several units. It is also possible to reduce the mass, dimensions, power consumption of the device while maintaining existing permissions and data transmission speeds.

Literature

1. Vasilenko GI Theory of recovery signals. On the reduction to the ideal device physics and technology. - M.: Owls. radio, 1979.

2. Mikhailov B.A. TO the choice of sampling frequency sufficient to restore the shape of a continuous signal. - Electronic equipment. Ser. 10. Microelectronic device 4 (46), 1984.

3. Michael L. Hilton, Bjorn D. Jawerth, 'ayan Sengupta. Compressing Still and Moving Images with Wavelets. - Multimedia Systems, Vol.2, No. 3, April 18,1994, p.7-11.

4. Image processing and digital filtering. Edited by T. Huang. - M.: Mir, 1979.

5. Patent RU No. 2141737.

6. atent RU # 2195014.

7. Korn G., Korn T. Handbook of mathematics for scientists and engineers. - M.: Nauka, 1974.

8. http://www.analog.com.

9. Digital and analog integrated circuits. Catalogue, part 1. Technical characteristics. - M.: Publishing house of the FSUE CCB "Dayton", 2002.

1. A device for processing signals containing serially connected frequency filter, the input of which is the input device, and discretization, characterized in that it additionally introduced block reduction to the perfect device intended for interpolation arriving at its input samples, determining weight basis of finite duration signals in the input signal based on the decomposition of interpolated samples in the Fourier series for orthogonalizing reactions frequency filter on the reference signals and to generate an output signal as a superposition of basis signals, depending on their weight in the input signal, and the input unit reduction to the perfect device connected with the output of discriminator, and the output is the output device.

2. A device for processing signals containing serially connected frequency filter, the input of which is the input device, and discretization, characterized in that it additionally connected in series block noise suppression designed for menshenina interference by normalizing the incoming input samples of the impulse elements with characteristics not relevant to the characteristics of possible output responses frequency of the filter, and block reduction to the perfect device intended for interpolation arriving at its input samples, determining weight basis of finite duration signals in the input signal based on the decomposition of interpolated samples in the Fourier series for orthogonalizing reactions frequency filter on the reference signals and to generate an output signal as a superposition of basis signals, depending on their weight in the input signal, and the input of the block noise reduction is connected with the output of discriminator, and the output unit reduction to the perfect device is the output device.

3. A device for processing signals containing serially connected frequency filter, the input of which is the input device, and discretization, characterized in that it additionally connected in series interpolator, an input connected to the output of discriminator, the determinant of the weights of the underlying finite duration signals in the input signal and the shaper output signal as a superposition of basis signals, depending on their weight in the input signal, the output of which is the output device.

4. A device for processing signals containing serially connected frequency filter, the input of which the CSO is the input device, and discretization, characterized in that it additionally connected in series, the node estimates the growth rate of a signal, an input connected to the output of discriminator, node normalization, the interpolator, the determinant of the weights of the underlying finite duration signals in the input signal and the shaper output signal as a superposition of basis signals, depending on their weight in the input signal, the output of which is an output device.

**Same patents:**

FIELD: information technologies.

SUBSTANCE: device comprises unit of input realization storage, clock oscillator, control unit, unit of useful signal extraction, unit of storage of five last values of useful component assessment, unit of approximation with polynom of the first degree, unit of approximation with polynom of the second degree, unit of output realization storage. In device end values of assessment are approximated with the help of method of least squares with polynom of the first or second degree, then produced equation of assessment is used to calculate values in forecast points.

EFFECT: forecasting measurement results on the basis of useful signal extraction without end effects, under conditions of limited a priori information about useful and accidental component.

1 dwg

FIELD: information technology.

SUBSTANCE: proposed invention relates to information measuring devices and can be used in computer engineering, in signal control and processing systems. The devices has a register for storing measurement results (1), delay unit (2), approximation unit (3), register for storing estimations (4), averaging unit (5), control unit (6), clock generator (7), and a comparator unit (8).

EFFECT: pickup of useful signal against a background of noise with minimisation of end effects, in conditions of insufficient prior information on statistical characteristics of adaptive noise and useful signal function given a single realisation of the measuring process.

4 dwg

FIELD: computer engineering.

SUBSTANCE: invention relates to digital computer engineering and can be used in digital signal processing systems for optimum nonlinear filtering. The device contains six units for generating matrix functions, three correctors, three units for generating difference, three units for generating sum, three delay lines, unit for generating and outputting prior data, and two units for calculating regularisation parametre.

EFFECT: increased accuracy of evaluating information process parametres in measuring systems.

8 dwg

FIELD: computer engineering.

SUBSTANCE: invention relates to computer engineering and can be used in systems for controlling and processing multi-dimensional signals. The device comprises a unit for storing input realisation 1, unit for calculating first coefficient 2, unit for calculating second coefficient 3, unit for calculating third coefficient 4, approximation unit 5, unit for evaluating useful component 6 and clock generator 7. Approximation of values of initial discrete realisation of the measured process is done through minimisation of the objective function, which is a sum of mean square deviations of two-dimensional evaluation of useful component from the initial two-dimensional realisation of measurement results.

EFFECT: separate two-dimensional evaluation of useful component given a single realisation of the measured process.

1 dwg

FIELD: physics; computer engineering.

SUBSTANCE: invention relates to computer engineering and can be used in control and signal processing systems. Technical outcome is achieved due to that, the device contains a unit for storing measurement results, commutators, interval division unit, random number generator, unit for eliminating associated values, ranking unit, storage register for random number samples, approximation units, subtracting units, remainder storage units, units for obtaining an ordered series on intervals, truncated sampling units, units for calculating mean-square deviation, multiplier units, coefficient storage register, coefficient evaluator, unit for setting false alarm probability, comparators, penalty storage units, arithmetic adder, threshold evaluator, comparator, penalty storage register, unit for eliminating anomalous measurements, storage register, delay unit and a clock pulse generator.

EFFECT: detection and elimination of anomalous measurements with a fixed value of false alarm probability.

1 dwg

FIELD: information technology.

SUBSTANCE: present invention relates to digital computer technology and can be used in systems for digital processing radio signals for optimum non-linear filtration. The device has blocks for generating matrix functions (4, 6, 10, 12), corrector units (2, 8), differential generating units (1, 7), summing units (3, 9), delay line (5, 11), unit for generating and output of priori data (13). The device also has a unit for calculating regularisation parametre (14), which is linked to the rest of the units.

EFFECT: more accurate evaluation of the information process in measuring systems.

8 dwg

FIELD: computer science.

SUBSTANCE: device has sum forming blocks, matrix functions forming block, difference forming block, delay lines, apriori data output block.

EFFECT: higher precision.

6 dwg

FIELD: computer science.

SUBSTANCE: device has sum forming blocks, matrix functions forming block, difference forming block, delay lines, apriori data output block.

EFFECT: higher precision.

6 dwg

FIELD: information technology.

SUBSTANCE: present invention relates to digital computer technology and can be used in systems for digital processing radio signals for optimum non-linear filtration. The device has blocks for generating matrix functions (4, 6, 10, 12), corrector units (2, 8), differential generating units (1, 7), summing units (3, 9), delay line (5, 11), unit for generating and output of priori data (13). The device also has a unit for calculating regularisation parametre (14), which is linked to the rest of the units.

EFFECT: more accurate evaluation of the information process in measuring systems.

8 dwg

FIELD: physics; computer engineering.

SUBSTANCE: invention relates to computer engineering and can be used in control and signal processing systems. Technical outcome is achieved due to that, the device contains a unit for storing measurement results, commutators, interval division unit, random number generator, unit for eliminating associated values, ranking unit, storage register for random number samples, approximation units, subtracting units, remainder storage units, units for obtaining an ordered series on intervals, truncated sampling units, units for calculating mean-square deviation, multiplier units, coefficient storage register, coefficient evaluator, unit for setting false alarm probability, comparators, penalty storage units, arithmetic adder, threshold evaluator, comparator, penalty storage register, unit for eliminating anomalous measurements, storage register, delay unit and a clock pulse generator.

EFFECT: detection and elimination of anomalous measurements with a fixed value of false alarm probability.

1 dwg

FIELD: computer engineering.

SUBSTANCE: invention relates to computer engineering and can be used in systems for controlling and processing multi-dimensional signals. The device comprises a unit for storing input realisation 1, unit for calculating first coefficient 2, unit for calculating second coefficient 3, unit for calculating third coefficient 4, approximation unit 5, unit for evaluating useful component 6 and clock generator 7. Approximation of values of initial discrete realisation of the measured process is done through minimisation of the objective function, which is a sum of mean square deviations of two-dimensional evaluation of useful component from the initial two-dimensional realisation of measurement results.

EFFECT: separate two-dimensional evaluation of useful component given a single realisation of the measured process.

1 dwg

FIELD: computer engineering.

SUBSTANCE: invention relates to digital computer engineering and can be used in digital signal processing systems for optimum nonlinear filtering. The device contains six units for generating matrix functions, three correctors, three units for generating difference, three units for generating sum, three delay lines, unit for generating and outputting prior data, and two units for calculating regularisation parametre.

EFFECT: increased accuracy of evaluating information process parametres in measuring systems.

8 dwg

FIELD: information technology.

SUBSTANCE: proposed invention relates to information measuring devices and can be used in computer engineering, in signal control and processing systems. The devices has a register for storing measurement results (1), delay unit (2), approximation unit (3), register for storing estimations (4), averaging unit (5), control unit (6), clock generator (7), and a comparator unit (8).

EFFECT: pickup of useful signal against a background of noise with minimisation of end effects, in conditions of insufficient prior information on statistical characteristics of adaptive noise and useful signal function given a single realisation of the measuring process.

4 dwg

FIELD: information technologies.

SUBSTANCE: device comprises unit of input realization storage, clock oscillator, control unit, unit of useful signal extraction, unit of storage of five last values of useful component assessment, unit of approximation with polynom of the first degree, unit of approximation with polynom of the second degree, unit of output realization storage. In device end values of assessment are approximated with the help of method of least squares with polynom of the first or second degree, then produced equation of assessment is used to calculate values in forecast points.

EFFECT: forecasting measurement results on the basis of useful signal extraction without end effects, under conditions of limited a priori information about useful and accidental component.

1 dwg

FIELD: information technologies.

SUBSTANCE: device comprises serially connected frequency filter, digitiser and unit of reduction to perfect instrument (RPI), intended for interpolation of counts supplied to its inlet, detection of weight of basic final duration of signals in inlet signal on the basis of interpolated counts decomposition into Fourier series by orthogonalised reactions of frequency filter into basic signals and for formation of outlet signal as a superposition of basic signals with account of their weight in inlet signal, besides versions of device include connection of noise suppression unit or serialy connected unit of signal growth speed assessment and normalisation unit between digitiser and RPI unit.

EFFECT: improved resolution and sensitivity to elements of signal, increased efficiency and simplification of device for signals processing.

4 cl, 12 dwg

FIELD: information technology.

SUBSTANCE: device has a unit for storing input realisation, switches, approximation units, estimation storage units, arithmetic adder, a unit for storing useful component estimates, a control unit, a delay unit, a clock-pulse generator, two units for breaking down into intervals, each having a random number generator, a unit for averaging related values, a ranging unit and a register for storing random number samples. The control unit has a shift register for sampling column random numbers, a shift register for sampling row random numbers, a delay unit for sampling column random numbers, a delay unit for sampling row random numbers, a counter and a unit for checking conditions.

EFFECT: two dimensional estimation of the useful component in conditions with insufficient prior information on statistical characteristics of additive noise and useful component function.

2 dwg

FIELD: physics.

SUBSTANCE: function codes are rounded-off to the nearest level and the obtained codes are stored. The optimality criterion code is calculated and stored. Starting with a certain initial number L of the function code, the direction of rounding-off this code is changed and the optimality criterion code is calculated. If the optimality criterion code falls, the changed value of the code is stored and a new value of the optimality criterion code is calculated and stored, otherwise the initial L-th function code and the initial optimality criterion code are stored, and calculation is moved on to the next number of the function code L+1, where it is checked whether the optimality criterion code falls in the same way as that when the L-th function code was changed. Further, the process is continued until the optimality criterion code does not fall in a sequence of n function codes, read from the code value in which the last fall in the optimality criterion code took place.

EFFECT: reduced absolute error in the amplitude of the reproduced sinusoidal signal.