RU2219658C2 - Device for compressing and recovering spectrum width of limited-frequency-band electric data signals - Google Patents

Abstract

FIELD: data transfer engineering; electrical communication systems. SUBSTANCE: device has first and second continuous-to-digital signal converters, first and second code-conversion units, first and second comparison and selection units, first through fifth data bases, first and second computing units, first and second digital-to-continuous signal converters, as well as first and second control units. Recurrent m-step algorithm used in device enables convolution of group of source signal counts into single group equivalent count pulse whose amplitude and polarity illustrate similar parameters of source count pulses of respective group. EFFECT: provision for compressing signal spectrum width and respectively reducing frequency band without slowing down signal transmission. 4 cl, 1 dwg

Description

 Description text in facsimile form (see graphic part) T

Claims (4)

1. A device for compressing and restoring the width of the spectrum of information electrical signals with a limited frequency band, comprising on the transmission side a first continuous signal converter with an upper frequency F ₁ into a sequence of discrete signals with clock intervals T ₁ = 1 / 2F ₁ , a first discrete signal converter with clock intervals T _2> T ₁ in the continuous signals with an upper frequency F ₂ <F ₁ and a first control unit, and on the receiving side - of the second continuous drive signal to the upper frequency F ₂ after ovatelnost digital signals with clock intervals of T ₂ = 1 / 2F _2, the second converter of digital signals with clock intervals T ₁ <T ₂ in the continuous signals with an upper frequency F ₁ and a second control device, characterized in that the apparatus the first and second recoding administered devices, the first, second, third, fourth and fifth databases, the first and second comparison and selection devices, the first and second computing devices, while on the transmission side to the output of the first continuous signal to discrete converter to supply a sequence of binary code combinations U ₁ (t) displaying discrete samples of input continuous signals u (t) and following at clock intervals T ₁ , the input of the first transcoding device is used to convert each group of n code combinations u ₁ (t) into group code combinations of CCM _j (j = 1, 2, 3, ...) and their subsequent conversion into code combinations of CC _CCji (j = 1, 2, 3, ...; i = 1, 2, .., m) supplied to the first input of the first device for comparison and selection, the second, third and fourth inputs of this device, designed for parallel supply of code combinations KK _d , each of which coincides with one of the possible code combinations KK _eji , and corresponding to each KK _d combinations {x _di ,, y _d2 } _d information numbers x _di and y _d2 , where d1 = 1, 2; d2 = 1, 2, ..., D / 2; d = 1, 2, ..., D, a D is equal to the total number of different code combinations of KC _eji with the largest number of binary digits, connected respectively to the first, second and third outputs of the first database, designed to store all KK _d and {x _d1 , y _d2 } d and sequentially outputting them according to the commands of the first control device to the first comparison and selection device, intended for sequential comparison of each combination of KK _eji with code combinations KK _d and, when coinciding with KK _eji, one of the combinations KK _d highlight the corresponding combination { x _d1 , y _d2 } _d to feed x _i and y _i in the format x _d1 and y _d2 from the first and second outputs of the first device for comparison and selection, respectively, to the fourth and fifth inputs of the first computing device, the first, second and third inputs of which are designed to supply normalizing coefficients C and k ₁ at one format of combinations of KK _eji or C and k ₂ with another format of combinations of KK _eji , connected to the outputs of the second database of the same name, designed to store the values of the coefficients C, k ₁ and k ₂ and output C and k ₁ or C and k ₂ on each stage with number i (i = 1, 2, ..., m) by the commands of the first device control in the first computing device, the second output of which is designed to supply intermediate values of N _i (i = 1, 2, ..., m-1) group samples GO _j (j = l, 2, 3, ...) , is connected to the sixth input of the same device, and its first output, designed to supply the final values of H _m GO _j , displaying each group of n code combinations u ₁ (t), in the form of pulses of one or another amplitude and polarity following with the clock intervals T ₂ > T ₁ connected to the input of the first Converter of discrete signals into continuous signals from the top frequency F ₂ <F ₁ , and on the receiving side to the output of the second continuous signal converter into discrete ones for supplying a sequence of pulse samples with clock intervals T ₂ = 1 / 2F ₂ corresponding to group samples of GO _j (j = 1, 2, 3, ...) with H * _m values, where the asterisk * hereinafter means a possible difference in the general case of the variables marked by it from similar values on the transmission side, the fifth input of the second computing device is connected, the first, second and third inputs of which are for feeding at each stage from No. i (i = m, m-1, ..., 1) by A team of the second control device normalizing coefficients C and k ₁ or C and k ₂ are connected with similar outputs third database identical to the second database, sixth, seventh and eighth inputs of the second computing device, intended for the parallel supply of data of the boundaries, the centers of all the used areas

and the corresponding values of _{i i} are connected to the outputs of the first, second, and third fourth databases, respectively, designed to store these data, as well as to output them at each stage with No. i (i = m, m-1, ..., 1) by commands of the second control device to the second computing device, in which the first output for supplying the values H * _i (i = m-1, ..., 1) is connected to the fourth input, and the second and third outputs for supplying information numbers x * _i and y * _i , (i = m, m-1, ..., 1) are connected to the inputs of the first and second of the second comparison device, respectively, and you a boron, the third, fourth and fifth inputs of which, for parallel supply by commands of the second control device, KK _d , x _d1 and y _{d2 are} connected respectively to the first, second and third outputs of the fifth database, identical to the first database, the second comparison and selection device is intended for sequential comparison of each combination of numbers {x * _i , y * _i } coming from the second computing device with combinations of numbers {x _d1 , y _d2 } _d coming simultaneously with code combinations KK _d from the fifth database, and when matching with { x * _i , y * _i } od one of the combinations {x _d1 , y _d2 } _d extracts the corresponding code combination KK _d , which coincides with one of the code combinations KK * _eji , to feed it from the output of the second comparison device and select the input of the second transcoding device for converting every m code KK * _eji combinations in the group of codewords GKK * _j, a last in the n binary codewords following a clock periods t ₁ and generating a sequence u * ₁ (t), supplied with the output of the second input device is recoded second transfigured of Tell digital signals into continuous signals with an upper frequency F _1. 2. The device according to claim 1, characterized in that the analog-to-digital converter (ADC) is used as the first converter of continuous signals to discrete, the low-pass filter (LPF) as the first converter of discrete signals to continuous, and the second converter of continuous signals to discrete - a device for sampling by time (UD) and as a second converter of discrete signals to continuous - a digital-to-analog converter (DAC). 3. The device according to claim 1, characterized in that in the first computing device for calculating the values of group samples of GO _j (j = 1, 2, 3, ...) an m-stage recurrence algorithm is used. H _i = x _i k [H _i-1 + С (1 + 2у _i )], i = 1, 2, ..., m, where H _i (i ≠ m) is the intermediate value of GO _j , H _m is the final value of GO _j , as Н ₀ (Н _i-1 for i = 1), you can use the value of one of the signal samples u (t) included in j -th group, or some verification number by which it is possible to verify the correctness of the restoration of u (t) on the receiving side, or other information numbers; x _i is a binary number of ± 1, which corresponds to one of the values of x _d1 , x _i displays the characters "1", "0" in the KC _eji and carries part of the information about the amplitude and (or) sign of the next reference signal or other information; k is a positive fractional decimal number (<1) that plays the role of a normalizing coefficient to limit the range of variation of the values of H _i ; C is the normalizing coefficient (positive decimal number), its purpose is to make the value [Н _i-1 + С (1 + 2у _i ] always positive; y _i is an integer or fractional decimal number that corresponds to one of the values of _d2 , y _i indicates one or another range of values | H _i | / k = [H _i-1 + C (1 + 2y _i )] (| H _i | - the modulus H _i ; the width of each region is Δ = 2C (1 + 2y _min ); the center of the region corresponds to the value of C (1 + 2y _i ); the number of regions is M and carries full or partial information about the parameters (sign and amplitude) of one or more samples of the original signal (or other information), y _i represents a codeword of r _i; bits, for each i-th stage of formation WASTE y _{j i} can take on one of m = 2 values ^ri s (position) different values of y _i by y _min to y _max are changed with the same pitch δ, for example, _i = {0, 1, 2, 3, 4, 5, 6, 7} (here δ = 1; M = 8; r _i = log ₂ 8 = 3) or for _i = {0.5; 2.5; 4.5; 6.5} (here δ = 2; M = 4; V _i = log ₂ 4 = 2); taking into account x _{i, the} combination {x _i , y _i } carries the information represented by r _1j = (r _i +1) binary digits; m is the number of stages of formation of GO. 4. The device according to claim 1, characterized in that in the second computing device for converting pulses corresponding to group samples ГО _j (j = 1, 2, 3, ...) with Н * _m values in the aggregate of values {x _i , for _i } * (i = m, m-1, ... 1) and H * _o , an m-stage recursive algorithm of four operations is applied at each i-th stage 1) fixing the value (| Н _i |) and the sign (x * _i ) H * _i (i = m, m-1, ..., 1); 2) the calculation of | H * _i | / k; 3) identification of the region to which the found value | H _i | / k belongs, the values of y _i and C (1 + 2y _i ) the average value of the region; 4) the definition of H * _i-1 = | H * _i | / k-C (1 + 2y * _i ), where all the above parameters coincide with the same parameters in paragraph 3 of the formula, and the asterisk * indicates a possible difference in the general case of marked her variables from similar values on the transmission side.

Images