Method and device for increasing communication channel capacity

FIELD: electronic engineering.

SUBSTANCE: method involves building unipolar pulses on each current modulating continuous information signal reading of or on each pulse or some continuous pulse sequence of modulating continuous information code group. The number of pulses, their duration, amplitude and time relations are selected from permissible approximation error of given spectral value and formed sequence parameters are modulated.

EFFECT: reduced inetrsymbol interference; high data transmission speed.

16 cl, 8 dwg

The invention relates to the field of telecommunications, mainly to the transfer of information using methods of pulse and digital modulation, and can be used to generate signals with predetermined spectral characteristics, providing a minimum level of intersymbol distortion in the communication channel.

Known methods of increasing the bandwidth of the communication channel, as a rule limited to the formation of a continuous signal for which the uncertainty relation satisfy certain criteria, minimizing intersymbol distortion and/or increase the bandwidth of communication channels [1, 2]. The mathematical description of these signals appears to be the normalized Hermite functions or spheroidal wave functions. Hardware implementation of these functions is difficult and does not provide variability changes in accordance with the characteristics of the communication channel.

There is a method of digital data transmission with minimal expansion of the spectrum [3], the closest to the proposed invention. It reduces intersymbol distortion and leads to the formation of a transmitted into the communication channel optimized pulse having such predskazanija that provide a consistent inverse characteristic postdetection filter lower h is the frequency. Make predskazanija in the transmitted signal to reduce inter-symbol distortion due some compensation is not ideal processing devices at the receiving side. However, forming the signal thus only partly solves the problem of narrowing the spectrum, so that the communication channel intersymbol distortion decrease is insignificant. In this sense, devices that implement the signals in the class of functions Hermite or spheroidal wave functions, solve a General problem, namely: minimize intersymbol distortion arising from the transfer of information over the communication channel. However, in any of these options generated signals have a complex shape, and therefore the noise and interference of the communication channel can substantially distort. Signals, having the shape close to rectangular, the most resistant to interference and noise of the communication channel. There was a fairly long flat top pulse requires under the same distortion levels greater power of interference and noise.

The main disadvantage of the method described in [3], is that insertion predskazanija in the transmitted signal to reduce inter-symbol distortion due compensation is not ideal processing devices at the receiving side. At the same time the parameters of the transmitted signal (in the article) do not directly take into account the characteristics of the communication channel, and so on BosniaHerzegovina distortion of the procedure used preemphasis not directly affected. The result is a negligible gain in increasing the capacity of the communication channel. In addition, insertion predskazanija directly related only to the characteristics postdoctoral lowpass filter, instead of the communication channel. This leads to the fact that the level of intersymbol distortion will depend on the type and parameters of the communication channel.

The known device for transmitting signals via the communication channels using different methods of pulse modulation. These devices adequately described, for example, in[4, 5, 6, 7, 8] and are known as the standard pulse modulation techniques. Each of the devices can be used as a prototype for a method of pulse modulation. The General lack of any of these devices, taken as a prototype for a type of modulation is used as the main storage medium rectangular pulse whose parameters (amplitude, duration, phase and frequency) are determined by the modulation type. However, regardless of the method of modulation, the spectral density of each pulse is described by a function of the form sin(ω )/ω that asymptotically decreases as 0(1/ω ), and has an oscillating character. This behavior of the spectral characteristics of the actuator is t to that rectangular pulses when passing through the communication channel are blurred, and in maginalised space of a signal, the so-called intersymbol distortion or interference. Given the known relationship between the effective duration of the rectangular pulse of the media and its upper frequency range (meaning the ratio of uncertainty) it is impossible to increase the speed of information transmission across a communication channel with a specific bandwidth without loss of quality of information transmission. Obvious at first glance, the approach to increase the bandwidth of the communication channel associated with a decrease in the pulse duration of the media is unacceptable. This is due to the increase in the level of intersymbol distortion, and, as a consequence, at the receiving side, the transmitted information will contain significant errors.

The technical result achieved by the proposed method of increasing bandwidth and device for its realization (options), is the formation of such composite pulse signal of a rectangular shape (definition of composite signal will give a bit later), which reduces intersymbol distortion compared to conventional, well-known devices that operate with a single rectangular pulses, and thereby help to improve / min net is ü transmission of information over the communication channel (or, what is the same, to increase throughput). In addition, compared with other methods of signal carriers, constructed, for example, on the basis of Hermite functions or spheroidal wave functions, there is a high noise immunity characteristic of rectangular pulses.

Thus, the technical result reached by the formation of signals with predetermined spectral characteristics, minimizing intersymbol distortion, which increases the bandwidth of the communication channel. In addition, the generated signals have a high noise immunity, because belong to the class of pulse signals having a rectangular shape.

Later in the application materials will be used the term composite signal. Under the composite signal is understood as a sequence of rectangular pulses representing or digital, or pulse-modulated signal, whose amplitude, pulse duration and temporal correlation between them is chosen in such a way as to get the desired spectral characteristics and/or to minimize intersymbol distortion.

This technical result is achieved by the fact that in the known method of increasing the bandwidth of communication channels, consisting in building digital and pulse the but-modulated signals with predetermined spectral characteristics, in accordance with the invention, for a given spectral characteristics form a composite signal in the form of a limited sequence of digital or pulse signals, the number of which, their duration, amplitude and temporal correlation between them is chosen by the allowable error of approximation to a given spectral characteristic in which the level of intersymbol distortion and/or expansion of the spectrum of the composite signal with at least stored in the generated sequence by using the methods of pulse modulation of the transmitted information.

In addition, the technical result is achieved by the fact that the number of pulses in the composite signal is chosen in the range from 2 to n, where n is the number of pulses in the sequence determined from the condition of minimal sufficiency for the formation of the composite signal with the allowable error of the approximation to a given spectral characteristics and preservation of the guard interval between the composite signals.

In addition, the technical result is achieved by the fact that the amplitude ratio between the pulses forming the composite signal, choose from a condition of formation of the spectrum of the composite signal that minimizes the expansion and/or the level of intersymbol distortion.

In addition to the above, the technical result is achieved by the fact that the time relationship between the pulses forming the composite signal, choose from a condition of formation of the spectrum of the composite signal that minimizes the expansion and/or the level of intersymbol distortion.

The technical result is also achieved by the fact that the amplitude of the pulses in the sequence, forming a composite signal which is proportional to the current level of the reference modulating the information signal by using the method of amplitude modulation.

In addition, the technical result is achieved by the fact that the duration of the pulses in the sequence, forming a composite signal which is proportional to the current level of the reference modulating the information signal using pulse width modulation.

In addition, the technical result is achieved by the fact that the temporal position of the composite signal proportional to the current level of the reference modulating the information signal p and the use of the method photoimpulse modulation.

The technical result is also achieved by the fact that temporal correlation between the pulses in the sequence, forming a composite signal which is proportional to the current level of the reference modulating the information signal by using the method of pulse-frequency modulation.

As well as the technical result is achieved by the fact that the duration of the pulses in the sequence, forming a composite signal which is proportional to the current number of continuously successive units in the information code sequence.

In addition, the technical result is achieved by the fact that the total duration of the sequence of pulses forming a composite signal using pulse code modulation proportional to the current number of continuously successive units in the information code group.

And, finally, the technical result is achieved by the fact that the composite signal has a total duration which is defined as the sum of the durations of the individual pulses of the sequence, forming a composite signal, and time intervals between them, and choose from a condition of maximum throughput of the communication channel, for a given acceptable level of inter-symbol distortion and the value of the guard interval between the adjacent composite signals.

The specified technical result is achieved by implementing the claimed method of the device, increasing the bandwidth of communication channels, and its options.

The device according to the first embodiment, different from the prototype that the output sample rate of the input information signal connected to the inputs of the device sample-hold and pulse shaper run, the output of which is connected to the inputs of n parallel connected shapers of pulses of a given duration, and the output of the first one through the first scaling device connected to the first input of the adder, and outputs the n-1 other shapers of pulses of predetermined duration through the corresponding n-1 delay devices and n-1 scaling of devices connected to the remaining n-1 inputs of the adder, the output of which is connected to the input of the control amplifier, a control input which is connected to the output of the sample-and-hold, and the control input of the sample-hold is connected to the output of the pulse shaper reset input of which is connected to the input of the n-th scaling device.

The device according to the second variant, different from the prototype that the output sample rate of the input information signal connected to the inputs of the device sample-hold and the formation of the of the motor pulse of a given duration, the output of which is connected to the input of the first scaling device, and inputs the remaining n-1 scaling of devices connected to the output of the pulse shaper predetermined duration through the corresponding n-1 delay devices, each of the n outputs of the scaling device connected to n inputs of the adder, the output of which is connected to the input of the control amplifier, a control input which is connected to the output of the sample-hold control input through which the forming device reset pulse is connected to the input of the n-th scaling device, the output of the controlled amplifier is connected to the input of the conversion voltage in a time interval.

The device according to the third variant, different from the prototype in that it additionally introduced forming device triggering pulse, the input of which receives the phase-frequency pulse-modulated information sequence, the output of which is connected to the inputs of n parallel connected shapers of pulses of a given duration, and the output of the first one through the first scaling device connected to the first input of the adder, and outputs the n-1 other shapers of pulses of predetermined duration through the corresponding n-1 delay devices and n-1 scaling of devices connected to the steel n-1 inputs of the adder, the output which is the output device.

The device according to the fourth variant, different from the prototype in that it additionally introduced forming device triggering pulse and the device conversion time interval the voltage at the input of which receives pulse code information sequence, the output device converting the time interval to a voltage connected to the input of the sample-hold, the output of which is connected to the control input of the control amplifier, the output of pulse shaper start connected through a first scaling device to the first input of the adder, and the remaining n-1 inputs of which are connected to the outputs of n-1 parallel scaling of devices whose inputs are through the respective delay device connected to the output of the device forming the triggering pulse, the output of the adder through a controlled amplifier connected to the Converter unit voltage during the time interval, the input of the n-th scaling device through an additional delay device connected to the input of pulse shaper reset, the output of which is connected to the control input of the sample-and-hold.

The technical result is achieved, therefore, by forming a composite signal in view of the final sequence of digital or pulse signals, including, their duration, amplitude and temporal correlation between them is chosen by the allowable error of approximation to a given spectral characteristic in which the level of intersymbol distortion and/or expansion of the spectrum of the composite signal is minimal. In this case, the total spectrum of pulse or digital signals may be closest to the spectra of the signals generated using the Hermite functions, or spheroidal wave functions, or the inverse characteristics of the communications channel. In a generated sequence, by using the methods of pulse modulation, the transmitted information is stored.

Consider in more detail the implementation of the proposed method on the example of the amplitude-modulation (PAM). Below, in the description of the specific devices that implement known methods of pulse and digital modulation, will be given the necessary explanations for each of these methods.

When transmitting information over the communication channel method the aim of the information component of the signal present in the pulse amplitudes, which we will denote as_q. We write the General expression for the signal in the form

where the current timing pulses taken at time τ_q.

Usually the pulse posledovatel the ability (1) synchronous, in the sense of that

τ_q=qT, q=0,± 1,± 2,...

At the receiving device receives a sequence

where

The sum in the expression (3) describes the error due to intersymbol distortion. It is due to the influence of neighboring pulses of the current pulse.

Note that rectangular pulses, the most common methods of transmission of information via communication channels. From consideration of the expression (1) seems to indicate that the reduction in the duration of carrying momentum can increase the speed of information transmission. However, this approach is unacceptable because of the limited bandwidth of the communication channel. The limitation of this type is taken into account in the form of the uncertainty relation, which for a rectangular pulse shape is

where Δ T is the effective duration of the pulse signal x(t), F_maxthe upper frequency in the spectrum of the rectangular pulse defined by power method at the specified level. As a rule, it is usually 90-95% of the total energy of the signal.

Inequality (4) is, in fact, a limitation on the speed of information transmission over the communication channel with bandwidth W.

The level of intersymbol distortion when used as a carrier informationframework pulses, will be significant. This is because the speed of the descending tail of the spectral density of each pulse asymptotically decreases as O(1/ω ) and has an oscillating character. The mentioned properties of the behavior of the spectral characteristics lead to the fact that a significant part of the pulse energy is outside the bandwidth of the communication channel. Mathematically, it is interpreted as follows. Spectral response is multiplied by a finite frequency window (which is essentially the frequency response of the communication channel). If this work is subjected to Fourier transform, we find that the rectangular pulse at the output of the communication channel is transformed into a pulse with a cinched front and rear fronts. Here the appropriate mathematical calculations, due to the popularity of this phenomenon.

Thus, to increase the bandwidth of the communication channel it is necessary to reduce the level of intersymbol distortion by generating such signals whose energy is outside of the F_maxminimum. This can be achieved, for example, the synthesis of pulses on the basis of Hermite functions or spheroidal wave functions.

In the present method describes a different approach, based on building a composite signal consisting of some nite consequently the particular rectangular pulses.

We will adhere to the following forms of presentation. All necessary calculations are going to spend on the basis of comparison of the properties of a single rectangular pulse and the proposed composite signal.

Figure 1 shows a typical and composite signals. The last one is a sequence (as an example) of the three individual pulses.

For the standard pulse (figa) the only variable parameter ensuring the optimization with respect to the characteristics of the communication channel and, therefore, the transmission speed is the duration of T_c. The spectral characteristics of the pulse decreases asymptotically as O(1/ω ) and is an oscillating function of the type sin(ω )/ω .

For the composite signal shown in figb, variable parameters are:

τ_i- shift between adjacent pulses Δ t_i(also variable parameter),

m_i- the zoom factor additional pulses to the left and to the right from basic (Central), and, finally,

n is the number of pulses in the composition.

The total duration of the complex composite signal is equal to T_cand it does not have to be the same length as the original a single rectangular pulse.

It should be noted that the simple one is CNY pulse of rectangular shape in communications technology is often associated in the time domain with a rectangular data window, which in the frequency domain corresponds to the kernel of convergence of the Dirichlet Fourier series. For composite signal scaling factors m, the number of pulses n and the shear ratio τ representing the composite signal, allow the same interpretation to assume that the so formed discrete lattice window data set characteristics in the time domain allows you to build a kernel convergence of Fourier series with specified properties.

We describe a mathematical model of the composite pulse signal in the time domain. For the description we will use the definition of rect functions and assume the following model properties:

signal is symmetrical relative to the zero reference time;

- composite signal continued in the area of negative time values.

Without loss of generality reasoning, value and_q(0) can be taken equal to 1.

Note that the duration of the composite signal, containing n=2k+1 pulses, equal (see figb)

Consider the special case n=1. Then

The spectrum of the composite signal degenerates into a spectrum of a single pulse

General case: n=2k+1. Composite signal in the time domain is described as

As you can see, a simple common factor and_kdescription (6) is generalized to an arbitrary reference amplitude.

Based on theorem on the spectrum is shifted in time of the pulse and producing a grouping of similar members, write

For a particular case when τ_i=0, m_i=1 (i=1, k-1), the modulus of the spectral density in 2k+1 times the modulus of the spectral density of a single pulse. The rate of decay of the spectral components is the same as for a single pulse.

On the other hand (alternative limiting case), if Δ t_i=Δ t and τ_i=τ and the choice of size τ =2π ω /(2k+1) the sum of the spectral density is equal to zero. At intermediate values of time offset, scaling factors and durations of the pulse components of the composite signal, the spectral density is defined as the geometric sum of the individual spectral components of the pulse.

Let us introduce the dimensionless scaling factor

. (i=0, 1, 2, 3,... )

If μ_i>1, then there is a compression pulse Δ t_iwith respect to the signal duration Δ t₁. If μ_i<1 is the stretching of the pulse, i.e. the change of its temporal scale of the ABA.

Taking account of the comments made, let us write the expression for the spectrum of the composite signal

or taking into account the Euler formula

Convert the product of trigonometric functions of the sum and take into account that the definition of μ_ishould μ₀=1 (since μ₀=Δ t₁/Δ t₁and that m₀=1, τ₀=0 for the Central composite pulse signal. Then the last expression can be written as

In the resulting expression, and hereinafter in the description, are used for negative indices, we will adopt the following definition and_-i≡and_i.

Then do the following. Multiply and divide each term of the sum on

.

In the end get

Expression (8) is a segment of trigonometrical series. Moreover, after bringing it to the form (9), where the basis functions are functions of the form

the obtained segment of the generalized Fourier series.

The system of basic functions (10) are orthogonal in the frequency interval (-∞ , ∞ ) and can be cast to an orthonormal system with appropriate selection of coefficients β_i. In effect con is knosti signal s(t) and the extremities of his energy, you can submit it in the form of a generalized Fourier series in the system of basic functions of the form (10). If the spectrum of the signal is limited (in this case, the bandwidth of the communication channel), using only a finite number of States of decomposition, it is possible to approximate the total spectrum of the composite signal, given, for example, in the form of some function or constraints. The function and requirements or restrictions can be set on the basis of the overall technical objectives of the synthesis signal with the desired properties. Of course, the replacement of some finite amount allows you to build only an approximate description of residually conditions or functions, i.e. with some error.

It is important to note that each term of the sum (9) in the system of basic functions of the form (10) is associated with a specific impulse represented in the composite signal. Therefore, by appropriate choice of variable parameters amount, such as m_ithat τ_ithat Δ t_iand n, we can construct the optimal segment number, which provides the desired accuracy of the approximation to the generated spectral characteristic of the composite signal, and to associate it with a certain set of rect functions in the time domain.

If the norms of the system of basic functions (10) to choose

then S(ω ) is a finite number of terms of the series in the basis of the form (10)

S(ω )=S_oc(ω ,T)+S_Osh(#x003C9; ),

where S_oc(ω ,T) is the spectral characteristic of the signal s(t) of duration So Here S_Osh(ω ) understand component, defined as the approximation error of the spectral characteristics.

Due to the orthogonality of basis functions spectra of S_oc(ω ,T) and S_Osh(ω ) do not overlap, and their energy, i.e. the squares of the norms, are

For absolute measure of the approximation errors can be taken, for example, a distance equal to the norm of the error signal, i.e. if W_S(ω ) - energy spectrum of the signal s(t), then by theorem Rayleigh

The formation of arbitrary spectral characteristics of the composite signal as described above can be implemented on the basis of two approaches. The first one mentioned above and is, in fact, to the problem of approximation of a given spectral characteristics with the norm of the error determined by the accuracy of the approximation. Next is the conversion of the obtained design parameters basic functions of the segment number in amplitude and time correlation of the composite signal.

The second approach consists in building a classical variational problem with constraints. It is necessary to find the optimal ratio between the m_ithat τ_ithat Δ t_ifor ∀ i∈ 1,n, t is the cue, to ensure the formation of a given spectral characteristics in the frequency domain.

The optimization criteria can be, again, the maximum error of the approximation to a given spectral characteristics or, say, the minimum energy spectral components lying within a certain upper frequency range (for example, it may be the upper frequency bandwidth of the communication channel).

In any case, both approaches solve the same task to minimize the level of intersymbol distortion and, consequently, increases the bandwidth of the communication channel when set on his characteristics.

Let us consider as example 1 formation of a composite signal consisting of three pulses (n=3 and fixed). Consider also that the composite signal is symmetrical relative to the zero reference time. Write

Take also the condition m₀=1, which does not reduce the generality of the result. Then

Variable settings Δ t₁that Δ t₂that τ₁, m₁.

We Dene the optimality criterion of the desired parameters Δ t₁that Δ t₂that τ₁, m₁based on the minimal energy of the spectrum, for example the frequency of the first zero of the main lobe of the spectral characteristics equal to ω ₀=2π .

The experiment showed that the values of variable parameters in the formulation and conditions of the following tasks: Δ t₁=0.511, Δ t₂=0.415, τ₁=0.671, m₁=0.237.

The spectral characteristic of the composite signal for these parameters is shown in figure 2. The mathematical modeling results show that even for this simple model was able to significantly reduce the level of spectral components outside the selected upper frequency spectrum. Note that the power spectral components of the composite signal is equal to 0.047. At the same time for a single pulse, it is the value of 0.088.

For comparability, the full power of the spectral components found in the range [0, 20 ω₀] (for numerical calculations is equivalent to ∞ ), was taken as 1.

Some comments on the choice of the system of constraints for variations Δ t₁that Δ t₂that τ₁, m₁. A variational problem based on the fact that we know the bandwidth of the communication channel. Thus, the criterion of minimization, such as the level of spectral components at some frequency, necessarily associated (or bound) to the upper cutoff frequency of the frequency characteristics of the communication channel.

In addition, the one-parameter solution of timizations tasks is not the best. Can be formulated and multivariable optimization problem. However, given the illustraty this example and that of the driven reasoning are part of the description of the application for the proposed method, we should stop.

Example 2. The description of the device implementing this method increases the bandwidth of the communication channel by using the method of amplitude modulation.

The technical result is achieved in that in a device for implementing the method of increasing the bandwidth of the communication channel by using the method of amplitude modulation that contains the device sample rate of the input information signal, the output of which is connected to the inputs of the device sample-hold and pulse shaper run, the output of which is connected to the inputs of n parallel connected shapers of pulses of a given duration, and the output of the first one through the first scaling device connected to the first input of the adder, and outputs the n-1 other shapers of pulses of predetermined duration through the corresponding n-1 delay devices and n-1 scaling of devices connected to the rest of the n-1 inputs of the adder, the output of which is connected to the input of the controlled amplifier, a control input which is connected to the output of the sample-storage is, moreover, the control input of the sample-hold is connected to the output of the pulse shaper reset input which is connected to the input of the n-th scaling device.

In the drawing figure 3 presents the block diagram of the device for implementing the method of increasing the bandwidth of the communication channel by using the method of amplitude modulation.

The device for implementing the method of increasing the bandwidth of the communication channel by using the method of amplitude-modulation device sample rate of the input information signal 1, the pulse shaper run 2, n shapers of pulses of desired duration 3, 4,... , n+2, n-1 delay devices n+3, n+4,... , 2n+1, n scale devices 2n+2, 2n+3,... , 3n+1, the pulse shaper reset 3n+2, the device sample-hold 3n+3, the adder 3n+4 and managed amplifier 3n+5.

The output of the controlled amplifier is an output device.

Consider in more detail the implementation of the proposed method increases the bandwidth of the communication channel by using the method of amplitude modulation.

The input analog information signal s(t) after the operation of sampling using the sampling device 1 is converted to the form

where s_qthe samples of the signal s(t), taken the e at time t _q=qT_disk. Here T_disk- period of the synchronous discretized sequence of pulses determined in accordance with theorem counts (Nyquist theorem). Due to the restrictions on the duration of the reference pulse associated with the finiteness of the bandwidth of the communication channel (see inequality (4)), its temporal extent is Δ t. Strictly speaking, there should be inequality, i.e. should not be exceeded this value.

The amplitude samples of the signal s_qstored with a device sample-hold 3n+3. The discretized signal is supplied to the forming device triggering pulse 2. The latter generates a triggering pulse on the leading edge of the reference signal s_q.

In order to simplify calculations and reduce the amount of formulae, further reasoning and description of conduct applies only to the current reference signal s_q.

Accept the condition that the middle of the current pulse count is at zero on the time axis, i.e. τ_q=0.

Given that the parameters of the communication channel is known, and is formulated and solved for the specific conditions of the optimization problem, then we know the following parameters generated composite signal, namely, n is the number of pulses in the composition, timing τ_ibetween each and the pulse and, accordingly, the scale factors m _ifor each of them.

Triggering pulse obtained at the output of pulse shaper run 2, synchronously starts n shapers of pulses of desired duration 3, 4,... , n+2. The output of each of these pulses is the same (for example, single) amplitude. Moreover, the pulse duration Δ t₁appears at the output of the described device (of course, in modified form after a number of transformations over him) immediately after counting s_k. Positional relation of the pulses of different duration and their number is strictly fixed on the basis of the results of the solution of optimization problems for composite signal. Further, the pulse duration Δ t₁from the output of the pulse shaper 3 is supplied to the scaling device 2n+2, which sets the desired value of the amplitude of this component of the composite signal. Thus, at the first input of the adder 3n+4 signal

It should be noted that this pulse is obtained at the output of the pulse shaper 2, as it is through the device with zero time delay, i.e. τ₀=0.

Similarly consider other channels of the formation of the pulse components of the composite signal so that the output of the adder 3n+4 are signal

The signal corresponding to the expression (15) is fed to a controlled amplifier 3n+5, the transmission coefficient which is supplied to its control input voltage level generated by the device sample-hold 3n+3. Thus, at the output of the controlled amplifier 3n+5 received the signal

The need to store the voltage level at the output of the sample-hold 3n+3 disappears immediately after the signal passed through the delay device 2n+1 with the greatest of all other devices delay the delay time (the channel formation of the last pulse in the composite signal). Therefore, the trailing edge of this signal is generated by the pulse shaper reset 3n+2 reset pulse for the device sample-hold 3n+3.

Note that the so formed composite signal has a significantly better spectral characteristics, focused on reducing inter-symbol distortion. Therefore, its duration can be selected smaller than the duration of a single pulse rectangular type, with a comparable level of intersymbol distortion in the communication channel. Considering the fact that the repetition period of the pulses is not the main determining factor affecting the value of the upper frequency range is, but only on the level of intersymbol distortion, so the repetition period of the composite signal is also reduced. So, the repetition period of the composite signal is compared with the repetition period of a single smaller. This allows, with the same characteristics of the communication channel, to increase the upper frequency spectrum of the transmitted signal, and thus further improve the throughput of the communication channel and on this side. As noted above, the main limitation of bandwidth of the communication channel are intersymbol interference. Taking into account the fact that the spectral characteristics of the composite signal consistent with the parameters of the communication channel, the level below, and this improves the throughput of the communication channel.

Example 3. Description of the device that implements this method increases the bandwidth of the communication channel by using the method of pulse-width modulation.

The technical result is achieved in that in a device for implementing the method of increasing the bandwidth of the communication channel by using the method of pulse-width modulation containing the device sample rate of the input information signal, the output of which is connected to the inputs of the device sample-hold and pulse shaper specified duration, the output catalogobject to the input of the first scaling device, and the inputs of the remaining n-1 scaling of devices connected to the output of the pulse shaper specified duration through the respective delay device, the outputs of each of the n scaling of devices connected to n inputs of the adder And the output of which is connected to the input of the control amplifier, a control input which is connected to the output of the sample-hold control input through which the forming device reset pulse is connected to the input of the n-th scaling device, the output of the controlled amplifier is connected to the input device converting the voltage in the time interval.

In the drawing figure 4 shows the block diagram of the device for implementing the method of increasing the bandwidth of the communication channel by using the method of pulse-width modulation.

The device for implementing the method of increasing the bandwidth of the communication channel by using the method of pulse-width modulation device sample rate of the input information signal 1, the pulse shaper specified duration 2, n-1 delay devices 3, 4,... , n+1, n scaling of device n+2, n+3,... 2n+1, the pulse shaper reset 2n+2, the device sample-hold 2n+3, 2n adder+4 controlled amplifier 2n+5, unit conversion voltage in the time interval 2n+6.

The output device is a conversion voltage in the time interval represents the output device.

We describe the implementation of the proposed method increases the bandwidth of the communication channel using pulse width modulation.

Previously we give some technical materials that will be used in the future. Their goal is confirmation of the fact that the implemented device eventually generates a sequence of pulses constituting the composite signal, and have properties similar to the sequence generated by the device when using amplitude modulation.

In General, when using the standard method of formation of the pulse sequence on the basis of the rectangular pulses of the type last described by the expression

where a_qthe amplitude of the current reference modulating the information signal; T is the repetition period of the timing and Δ t - duration modulated pulse when the value of the information signal and_q=1. Media in this method of modulation is the duration of the current pulse. As for amplitude modulation, the spectral characteristic of each pulse sequence of the type (17) is described by a function of the form sin(ω )/ω . In addition, the spectrum width depends on the level of the modulating signal is La, while for amplitude-modulation of this dependence is not.

As in the previous case, using amplitude-modulation signal intersymbol distortion in the communication channel arise from the interaction of the "tails" of the spectral characteristics, which also limits the speed of information transfer.

It is obvious that this method of modulation optimization of the spectral characteristics of the modulated information of the pulse is also necessary. Suppose q reference pulse modulated pulse in the form of a composite signal, shown in figure 5. For clarity and simplicity of the calculations used in the model: (a_q=1 and symmetric with respect to the zero reference time. Thus, the composite signal for the q count is

The expression for the spectrum of the composite signal, taking into account theorem on the spectrum is shifted in time of the pulse is

or going from work trigonometric functions of the sum and given that m₀=1, we write

The resulting expression is in some sense similar to we derived earlier expressions for the case of amplitude modulation. Multiplying and dividing each term of the sum by a factor of

get

The obtained segment of trigonometric series of the form (20) is essentially the same segment number, described by expression (9).

Thus, it is also the cut of the generalized Fourier series, with all the ensuing properties mentioned earlier. Variational or approximation tasks are based on the same principles that have been formulated previously.

We present a description of the implementation of the proposed method increases the bandwidth of the communication channel using pulse width modulation.

The input analog information signal s(t) after the operation of sampling using the sampling device 1, is converted to the form (14).

The amplitude samples of the signal s_qstored with a device sample-hold 2n+3. The discretized signal is applied to the pulse shaper 2 duration Δ t₁that generates a triggering pulse on the leading edge of the reference signal s_q.

In order to simplify calculations and reduce the amount of formulae further reasoning and description as well as for example 2, conduct applies only to the current reference signal s_q. Accept the condition that the middle of the current pulse count is at zero on the time axis, i.e. τ_q

Given that the parameters of the communication channel is known, and is formulated and solved for the specific conditions of the optimization problem, then we know the following parameters generated composite signal, namely, n is the number of pulses in the composition, timing τ_ibetween each pulse and, accordingly, the scale factors m_ifor each pulse.

The amplitude of the output signal of the pulse shaper 2 duration Δ t₁random and fixed (conventionally it can be taken equal to unity) and is supplied through the n-1 delay devices 3, 4,... , n+1, the corresponding n-1 scaling of device n+3, n+4,... , 2n+1. Moreover, the scaling device n+2 is connected to an input directly to the output of the pulse shaper 2 duration Δ t₁.

As in the previous example, we known positional relation, i.e. the time delay of each pulse in the song τ_iand scale factors m_iand their number n. Note that we formulate a variational problem in this case is slightly different according to the number of variable parameters. Thus, the time duration of the pulse components of the composite signal, the same and equal to Δ t₁.

The n outputs of the scaling of device n+2, n+3,... ,2n+1 is connected to n inputs of the adder 2n+4. On it in the course of a signal is generated

served on a managed amplifier 2n+5, the transmission coefficient is determined by the level of the signal present at its control input. This signal is generated at the output of the sample-hold 2n+3. The device sample-hold 2n+3 is controlled by the pulse shaper reset 2n+2, whose input is connected to the output of the delay device n+1. Thus, at the output of the controlled amplifier 2n+5 is formed sequence

supplied to the Converter unit voltage during the time interval 2n+6, the output of which have

Thus, the received sequence representing the composite signal with the required properties.

At the end of the pulse duration Δ t₁the delay device n+1 is formed by a reset pulse on the trailing edge of the pulse shaper reset 2n+2. The reset signal resets the device sample-hold 2n+3, and thus it is prepared for formation of the next composite signal.

Conclusions for the considered method of pulse modulation are similar to the findings of for example 2.

Example 4. Description of the device that implements this method increases the bandwidth of the communication channel when using the ecogov phase-pulse-frequency modulation.

The technical result is achieved in that in a device for implementing the method of increasing the bandwidth of the communication channel by using the methods of phase-pulse-frequency modulation, characterized in that it additionally introduced forming device triggering pulse, the input of which receives the phase - frequency pulse-modulated information sequence, the output of which is connected to the inputs of n parallel connected shapers of pulses of a given duration, and the output of the first of them is connected to the first input of the adder, and outputs the n-1 other shapers of pulses of predetermined duration through the corresponding n-1 delay devices connected to the remaining n-1 inputs adder whose output is the output device.

Drawing 6 shows the block diagram of the device for implementing the method of increasing the bandwidth of the communication channel by using the methods of phase-pulse-frequency modulation.

The device for implementing the method of increasing the bandwidth of the communication channel by using the methods of phase-pulse-frequency modulation device forming pulse start 1, n shapers of pulses of desired duration 2, 3, ... ,n+1, n-1 delay devices n+2, n+3,... ,2n, 2n adder+1.

The output of the adder is the output device is Ista.

Known methods of forming phase-pulse or pulse-frequency signals are reduced to the formation sequence of the form

where a_qthe amplitude of the current reference modulating the information signal; T is the repetition period of the pulses in synchronous sequence when the value of the information signal and_q=1. For pulse modulation techniques-data carrier is periodic synchronous sequence of rectangular pulses with distinctive characteristics typical of those or other types of modulation. For example, in the case where a_q=1 (q=1, 2,... ), there is a sequence of period T, or, in other words, each subsequent pulse is delayed by the same time, so that its phase relative to the clock time is fixed and is equal to zero. In the same case, if you use a phase-pulse modulation, the phase of the rectangular pulse in the sequence is changed according to the modulating information signal.

For pulse-frequency modulation position, the position of the pulses in the sequence also depends on the level of the modulating signal. Moreover, the frequency of the pulses in the sequence is determined by the level of the information signal.

Combining both methods mod the ablation in a single device is based on the full identity of the processing of their signals during formation of the composite pulse.

Each individual pulse sequence (24) has a spectral characteristic that is described by a function of the form sin(ω )/ω . Differences in the methods used modulation in relation to the above leads to the fact that all the phase - and frequency-modulated pulses in the transmitted sequence absolutely identical to the description, as in the time domain and frequency. I.e. the spectral characteristics of each individual pulse is the same. However, the methods themselves modulation, leading to a positional offset of the pulses in the sequence, proportional to their modulation parameter, all the same, as and for the above methods lead to the occurrence of inter-symbol distortion. Their level is, in fact, limits the bandwidth of the communication channel at a given transmission quality information. Thus, there is a need for optimizing the spectral characteristics of the transmitted signal.

Below, we will not give a detailed mathematical description of the spectral sequence characteristics phase-frequency-modulated signals. In many ways it is same as discussed above, the spectral characteristics of each individual pulse in his description correspond to the description of the spectral characteristics of the conventional rectangular pulse with according to the relevant specifications.

Replace the current reference phase-frequency pulse signal of the composite sequence.

We adopt the following model properties of the composite signal for the q reference frame. First, the positional offset of the Central pulse on the time axis is equal to a_qand itself composite pulse sequence symmetric about the point t=a_q. Thus, the composite signal is described by the expression

When the spectrum of the composite signal should be aware that the offset sequence on the time axis by the value t=a_qonly leads to the change of the phase spectrum, while the amplitude spectrum remains the same as for momentum, the middle of which is the zero of the time axis. Therefore, to simplify the mathematical expressions slip composite sequence by the value t=a_q. So, we write

The spectrum of the composite signal, taking into account theorem on the spectrum is shifted in time and momentum properties of symmetry after some transformations has the form

where μ_i=Δ t₁/Δ t_i+1(I=0, 1, 2,... ) - dimensionless scaling factor. The expression (26) is largely similar to (8), except that all mysteryous the e coefficients m _i=1 (i=1, k-1). After conversion works trigonometric functions of the sum and given that it μ₀=1, τ₀=0, the relation (26) we can rewrite as:

Multiply and divide each term of the sum on

get

Obviously, in this case, the period of the trigonometric series in the system of basic functions (10).

The choice of the parameters n=2k+1, Δ t_ithat τ_iin a similar way as for the above examples.

We describe the implementation of the proposed method increases the bandwidth of the communication channel using the phase, pulse-frequency modulation.

Input device formation triggering pulse 1 arrives phase-frequency-modulated sequence. The forming device triggering pulse 1 generates a pulse on the leading edge of the received signal and starts n shapers of pulses of desired duration 2, 3,... n+1. At the output of the shaping pulse 2 pulse is generated duration Δ t₁fixed (for example, single) amplitude and which is fed to the first input of the adder 2n+1. The other n-1 devices forming pulses 3, 4,... , n+1 generate pulses Δ t₂that Δ t_{3 and so Also with a fixed amplitude equal to the amplitude at the output of the impulse formation 2.}

_{Mentioned n-1 pulses through the respective n-1 delay devices n+2, n+3,... ,2n, proceed on the remaining n-1 inputs of the adder 2n+1. At its output, a signal is generated}

_{In the General case, when the displacement of the Central pulse in the composite sequence arbitrarily, in the expression (28) the numerator is modified as: t(aq+τi-1).}

Conclusions regarding the thus obtained composite signal would remain the same as was done previously in the examples of implementation of those or other devices forming the composite signal for the corresponding modulation techniques.

Example 5. Description of the device that implements this method increases the bandwidth of the communication channel using the method of pulse code modulation.

The technical result is achieved in that in a device for implementing the method of increasing the bandwidth of the communication channel using the method of pulse code modulation, characterized in that it additionally introduced forming device triggering pulse and the device conversion time interval the voltage at the input of which receives pulse code information consequently is here, the output device converting the time interval to a voltage connected to the input of the sample-hold, the output of which is connected to the control input of the control amplifier, the output of pulse shaper start connected through a first scaling device to the first input of the adder, and the remaining n-1 inputs of which are connected to the outputs of n-1 parallel scaling of devices whose inputs are through the respective delay device connected to the output of formation of the triggering pulse, the output of the adder through a controlled amplifier connected to the Converter unit voltage during the time interval, the input of the n-th scaling device through an additional delay device connected to the input of pulse shaper reset the output of which is connected to the control input of the sample-and-hold.

In the drawing of Fig. 7 presents a block diagram of a device for implementing a method of increasing the bandwidth of the communication channel using the method of pulse code modulation.

The device for implementing the method of increasing the bandwidth of the communication channel when using the pulse code modulation device forming pulse start 1, n-1 delay devices 2, 3,... , n, n scaling of device n+1, n+2,... , 2n, additionally the e delay device 2n+1, the pulse shaper reset 2n+2, the device conversion time interval in voltage 2n+3, the device sample-hold 2n+4, 2n adder+5 controlled amplifier 2n+6, unit conversion voltage in the time interval 2n+7.

The output of voltage conversion time interval is the output device.

The input analog signal using the coding device is converted into a sequence of binary digits transmitted at the rate of one binary bit every T seconds. The combination of 1 and 0 form a code group is received at the input device to implement the method of increasing the bandwidth of the communication channel using the method of pulse code modulation. Thus,

In equation (29), the following notation: Δ t - pulse duration corresponding to 1 one digit code; T - period of the sequence of binary digits (regardless of the value of this discharge, whether 0 or 1); K(t) - function-indicator

Because the amplitude of the digital signal does not play a significant value, then let's make it equal to 1. The carrier of the transmitted information is the presence or absence of a digital pulse. Take into account that each unit in the discharge of code corresponds to a rectangular them is the pulse, the spectrum of which is described by a function of the form sin(ω )/ω ). The effective width of the spectrum, defined for example by frequency band from zero to the first intersection of the spectral function with the axis of the frequencies depends on the number of continuously following single digits code. The greatest width of the spectrum is responsible only one code from the previous and subsequent values of zero digits code. Obviously, there are intersymbol distortion (sometimes referred to as maineline), the level of which is the limitation of the speed of information transmission over the communication channel.

The generated composite signal replaces the continuous rectangular pulse corresponding to 1 in the category code groups, a set of pulses of different duration and shifted relative to each other at a specified time interval. Given the fact that the binary bit code is equal to 1, can appear at any time, you will build a model of the composite signal based on the accepted traditional schemes. Thus, the composite signal is symmetric with respect to zero of the time axis, and the middle of the pulse is at the point t=0. Let's write the following:

It is easy to see that this sequence is identical to sequence of pulses of the composite signal to Shir the IDT-width modulation (see the expression (18)). This allows you to record directly the final expression for the spectrum of the composite signal

So, the newly obtained segment of trigonometric series in the system of basic functions (10). Thus, the basic properties, its application and methods of determining optimal parameters described above.

We present a description of the implementation of the device according to the method of increasing the bandwidth of the communication channel using pulse code modulation.

The input sequence of binary numbers supplied to the forming device triggering pulse 1 and device conversion time interval in voltage 2n+3.

The forming device triggering pulse 1 generates a pulse Δ t on the trailing edge of the pulse corresponding to 1 bit or a continuous sequence of such units. The generated pulse Δ t through n+1 of the scaling device is fed to the first input of the adder 2n+5. On the remaining n-1 inputs of the adder 2n+5 receives signals from n-1 scaling of device n+2, n+3,... , 2n whose inputs are connected to the corresponding outputs of n-1 delay devices 2, 3,... , n. The inputs of the latter is connected to the output of the shaping pulse start 1. At the output of the adder 2n+5, a signal is generated

which goes to the input of the control amplifier 2n+6. At its control input voltage from the device sample-hold 2n+4. The voltage at the output of the sample-hold 2n+4 proportion to the length of a continuous sequence of units in the digits of the code group. This voltage is generated on the output device converting the time interval to voltage 2n+3, Thus, the output of the controlled amplifier 2n+6 received the signal

where γ - dimensional large-scale conversion factor of the Converter of the time interval in voltage for single digits binary code received in the course of time, when K(t)=1.

The output of the controlled amplifier 2n+6 (32) is supplied to the Converter unit voltage during the time interval 2n+7, and its output is formed by a sequence of the form

and that is precisely what was required to get.

The cycle of formation of composite sequence is finished, and the pulse shaper reset 2n+2 clears the device sample-hold 2n+4. The start pulse shaper reset 2n+3 is on the trailing edge formed by the last pulse in the sequence of the composite signal appearing at the output of delay device n. With formirovanii at the output of delay device n the signal is sent to the accessory delay 2n+1. The time delay which must be at least the time of the conversion voltage in the time interval unit conversion voltage in the time interval 2n+7.

The main findings, criteria and methods for providing a set of optimal parameters of the composite signal method for pulse code modulation did not differ from the above.

Literature

1. Aigulin, Vpechatliv. Compactly supported functions in physics and engineering. M., Nauka, 1971.

2. Linex. Theory of signals. M, Owls. Radio, 1974.

3. Spectral efficient system digital frequency modulation. IPC: N 03 1/06, H 03 D 1/06, H 03 K 3/013, PCT(WO). Inventions of the world. Vol. 108, No. 2, 1992.

4. Whmsonic, Humongous. Nonlinear control systems with a frequency and pulse-width modulation. Kiev, 1970.

5. Annavasal, Affain. Fundamentals of theory and calculation of information-measuring systems. M., engineering, 1980.

6. UMIAT. Circuits, signals, systems. M., Mir, 1988.

7. Ivimy, Vagitus. Fundamentals of information theory and coding. Kiev, high school, 1986.

8. Shibasoku. Radio circuits and signals. M., High school, 1983.

1. The method of generating signals with a predetermined spectral characteristic, namely, that for each current reference continuous modulating the information signal or for each or continuous placentas the activity of pulses modulating information code groups form a unipolar pulses, including, their duration, amplitude and timing of choose a valid approximation errors for a given spectral characterization and modulate the parameters of the generated sequence.

2. The method according to claim 1, characterized in that the number of pulses in a bounded sequence of unipolar pulses is chosen in the range from 2 to n, where n is the number of pulses in the sequence determined from the condition of minimal sufficiency, providing a margin of error of the approximation to a given spectral characterization and preservation of the guard interval is formed between the limited sequences.

3. The method according to claim 1, characterized in that the amplitude ratio between the pulses forming a limited sequence of unipolar pulses, which are selected from the conditions of formation of its spectrum that corresponds to the given optimization criterion.

4. The method according to claim 1, characterized in that the pulse duration limited sequence of unipolar pulses are selected from the conditions of formation of its spectrum that corresponds to the given optimization criterion.

5. The method according to claim 1, characterized in that the time relationship between the pulses forming a limited sequence of unipolar pulses, which are selected from the conditions of formation of its spectrum, the CTE is committed to a given optimization criterion.

6. The method according to claim 1 or 3, characterized in that the amplitude of the pulses bounded sequence of unipolar pulses is proportional to the current level of the reference modulating the information signal by using the method of amplitude modulation.

7. The method according to claim 1 or 4, characterized in that the pulse duration in a bounded sequence of unipolar pulses is proportional to the current level of the reference modulating the information signal using pulse width modulation.

8. The method according to claim 1, characterized in that the temporary regulation bounded sequence of unipolar pulses proportional to the current level of the reference modulating the information signal by using the method photoimpulse modulation.

9. The method according to claim 1 or 5, characterized in that the timing between pulses in a bounded sequence of unipolar pulses is proportional to the current level of the reference modulating the information signal by using the method of pulse-frequency modulation.

10. The method according to claim 1 or 4, characterized in that the pulse duration in a bounded sequence of unipolar pulses proportional to the number of continuously successive units of information code group.

11. The method according to claim 1 and and 4, characterized in that the total duration of a bounded sequence of unipolar pulses using pulse code modulation proportional to the current number of continuously successive units in the information code group.

12. The method according to claim 4 or 5, characterized in that limited sequence of unipolar pulses of total duration, defined as the sum of the durations of the individual pulses of the sequence and time intervals between them, and choose from a condition of maximum speed of information transmission over the communication channel, for a given acceptable level of inter-symbol distortion and the value of the guard interval is formed between limited unipolar sequences.

13. A device for generating signals with predetermined spectral characteristics when using the amplitude-modulation containing the device sample rate of the input information signal, characterized in that the output sample rate of the input information signal connected to the inputs of the device sample-hold and pulse shaper run, the output of which is connected to the inputs of n parallel connected shapers of pulses of a given duration, and the output of the first one through the first scale is the dominant device connected to the first input of the adder, and outputs the n-1 other shapers of pulses of predetermined duration through the corresponding n-1 delay devices and n-1 scaling of devices connected to the remaining n-1 inputs of the adder, the output of which is connected to the input of the control amplifier, a control input which is connected to the output of the sample-and-hold, and the control input of the sampling device storage is connected to the output of the pulse shaper reset on the trailing edge, the inlet of which is connected to the input of the n-th scaling device.

14. A device for generating signals with predetermined spectral characteristics when using the method of pulse-width modulation containing the device sample rate of the input information signal, characterized in that the output sample rate of the input information signal connected to the inputs of the device sample-hold and pulse shaper specified duration, the output of which is connected to the input of the first scaling device, and inputs the remaining n-1 scaling of devices connected to the output of the pulse shaper predetermined duration through the corresponding n-1 delay devices, each of the n outputs of the scaling device connected to n inputs of the adder, the output of which is connected to the input of the control amplifier, a control input which connection is Chen to the output of the sample-hold the control input of which is through the formation of a reset pulse on the trailing edge connected to the input of the n-th scaling device, the output of the controlled amplifier is connected to the input of the conversion voltage in a time interval.

15. A device for generating signals with predetermined spectral characteristics when using the methods of the phase, pulse-frequency modulation, characterized in that it additionally introduced forming device triggering pulse, the input of which receives the phase, pulse-frequency modulated information sequence, the output of which is connected to the inputs of n parallel connected shapers of pulses of a given duration, and the output of the first one through the first scaling device connected to the first input of the adder, and outputs the n-1 other shapers of pulses of predetermined duration through the corresponding n-1 delay devices and n-1 scaling of devices connected to the rest of n-1 inputs of an adder whose output is the output device.

16. A device for generating signals with predetermined spectral characteristics using the method of pulse code modulation, characterized in that it additionally introduced forming device triggering pulse and the device pre is obrazovaniya time interval in voltage, the inputs which receives pulse code information sequence, the output device converting the time interval to a voltage connected to the input of the sample-hold, the output of which is connected to the control input of the control amplifier, the output of pulse shaper start connected through a first scaling device to the first input of the adder, and the remaining n-1 inputs of which are connected to the outputs of n-1 parallel scaling of devices whose inputs are through the respective delay device connected to the output of formation of the triggering pulse, the output of the adder through a controlled amplifier connected to the Converter unit voltage during the time interval, the input of the n-th scaling device through additional delay device connected to the input of pulse shaper reset on the trailing edge, the output of which is connected to the control input of the sample-and-hold.