JP4765065B2 - Chaotic coded modulation demodulation method - Google Patents

Description

    The present invention relates to a coded modulation demodulation method in which information is transmitted by coded modulation between a transmission side and a reception side, and in particular, even when a received signal deteriorates due to noise in a communication channel in digital communication. The present invention relates to a chaotic coded modulation / demodulation method that can perform high-quality transmission and cannot be easily decoded by others.

  In the fields of mobile wireless communication and wireless LAN, the demand for higher speed and higher efficiency is increasing due to the recent spread of multimedia communication. However, in mobile communication or the like, intersymbol interference occurs due to multipath, and the propagation path environment frequently changes, so it is necessary to establish high-quality communication in a poor environment. On the other hand, in recent years, electronic commerce systems using wireless communication terminals are gradually spreading, and ensuring security in communication and realizing highly confidential communication are very important.

  By the way, in the field of channel coding technology, in order to realize high-quality communication in a worse environment, modulation methods and coding technologies have been improved. Among them, the turbo code is a technique for realizing high quality transmission approaching the Shannon limit. In this method, error correction capability is greatly improved by performing iterative decoding via interleaver for parallel connection of codes. In the field of encryption technology, a number of methods that exhibit excellent characteristics have been proposed (see Patent Document 1). Furthermore, the present inventor has proposed a block coded modulation method using a chaotic equation (see Non-Patent Document 1).

JP 2001-326631 A Eiji Okamoto, “A Study on Coded Modulation Using Chaotic Equations,” Shingaku Techniques, RCS2001-307, pp. 159-164, Mar. 2002

  However, since the turbo code described above requires a relatively large interleaver and iterative calculation, there is a problem that the decoding result cannot be extracted sequentially. On the other hand, the conventional encryption technology is used in a concept different from encoding, and the encoding technology for high-quality transmission and the encryption technology for enhancing confidentiality are processed separately. Therefore, there is a problem that the calculation scale has increased.

  The problem to be solved is that relatively good transmission error rate characteristics can be obtained even with low received signal quality, and that sequential decoding is possible, and that the communication signal sequence is close to noise, so that others can easily decode the communication contents. It is to provide a chaotic coded modulation system that is incapable of being concealed.

  Hereinafter, each means suitable for solving the above-described problems will be described with additional effects as necessary.

1. A coded modulation demodulation method for transmitting information by coded modulation between a transmitting side and a receiving side,
The sender side
A transmission information bit sequence is input to a chaos generator to generate a coded signal sequence, and a chaos transmission signal sequence generation step for transmitting to the reception side as a chaos transmission signal sequence of a coded modulation scheme is provided,
The receiving side
An estimated transmission sequence generating step for generating an estimated transmission sequence;
An estimated transmission signal sequence generating step for generating an estimated transmission signal sequence by the same processing as the chaotic transmission signal sequence generating step on the transmission side with the estimated transmission sequence as an input;
A decoding step of calculating an error between the received signal sequence received from the transmitting side and the estimated transmission signal sequence, and outputting an estimated transmission sequence that gives the minimum error as a decoding result of the received signal sequence , and
The chaotic transmission signal sequence generation step on the transmission side includes:
An input-side operation step of generating an input signal sequence by performing a predetermined input-side operation with the transmission information bit string and the fed back chaotic sequence as inputs; and
A chaos sequence generation step of generating a chaos sequence by the chaos generator with the input signal sequence as an input; and
An output side computation step for generating the chaotic transmission signal sequence by performing a predetermined output side computation with the chaos sequence as an input; and
With
The estimated transmission signal sequence generation step on the receiving side includes:
An input side calculation step of generating an estimated input signal sequence by performing the same input side calculation as the transmission side with the estimated transmission sequence and the feedback estimated chaotic sequence as inputs,
An estimated chaos sequence generating step for generating an estimated chaos sequence by the same chaos generator as the transmission side with the estimated input signal sequence as an input;
A chaotic coded modulation / demodulation method comprising: an output-side operation step of generating an estimated transmission signal sequence by performing the same output-side operation as that of the transmitting side with the estimated chaotic sequence as an input .
According to the means 1, in the transmission side, the chaos transmission signal sequence generation step inputs the transmission information bit string to the chaos generator to generate the encoded signal sequence, and transmits it to the reception side as the chaotic transmission signal sequence of the coded modulation scheme. Then, on the receiving side, the estimated transmission sequence generation step generates an estimated transmission sequence, and the estimated transmission signal sequence generation step receives the estimated transmission sequence as an input and performs the same processing as the transmission side chaotic transmission signal sequence generation step. The decoding step calculates an error between the received signal sequence received from the transmitting side and the estimated transmission signal sequence, and outputs an estimated transmission sequence that gives the minimum error as a decoding result of the received signal sequence. Therefore, by performing analog coding using a chaotic sequence, that is, coding by a signal waveform, it is possible to achieve both good transmission characteristics and transmission confidentiality. That is, relatively good transmission error rate characteristics can be obtained even with low received signal quality, and sequential decoding is possible. Furthermore, since the communication signal sequence varies randomly due to chaos, it is close to noise, and it is possible to transmit information with excellent confidentiality that cannot be easily decoded by other people. In other words, because the transmission signal changes in a pseudo-noise manner by using a chaotic sequence, the transmission information is not clear in the signal itself, and all parameters in the chaotic transmission signal sequence generation step used on the transmission side and the reception side must be grasped. The secrecy of the transmission is realized because it is almost impossible for others to decode due to the uncorrelatedness of the chaotic sequence. Any type of chaos generator may be used, and a plurality of systems may be mixed. The degree of freedom of calculation in the encoding process is high, and the encoding rate can be set freely.
In particular, on the transmission side, the input side computation step generates the input signal sequence by performing a predetermined input side computation with the transmission information bit string and the fed back chaotic sequence as inputs, and the chaos sequence generation step inputs the input signal sequence. A chaos sequence is generated by a chaos generator, and an output side calculation step generates a chaos transmission signal sequence by performing a predetermined output side calculation with the chaos sequence as an input. On the other hand, on the receiving side, the estimated transmission sequence generation step generates an estimated transmission sequence, and the input side calculation step performs the same input side calculation as the transmission side with the estimated transmission sequence and the feedback estimated chaotic sequence as inputs. An estimated input signal sequence is generated, an estimated chaos sequence generation step receives the estimated input signal sequence as an input, generates an estimated chaos sequence by the same chaos generator as the transmission side, and an output side calculation step receives the estimated chaos sequence as an input side. The estimated transmission signal sequence is generated by performing the same output side calculation as in step (b), and the decoding step calculates the error between the received signal sequence received from the transmission side and the estimated transmission signal sequence, and calculates the estimated transmission sequence that gives the minimum error. Output as a decoding result of the received signal sequence. As the input side calculation, any calculation may be used as long as it is within the convergence range of the chaos generator, and as the output side calculation, for example, a calculation that limits the maximum amplitude or the like may be used. .

2. The predetermined input-side operation f is expressed by the formula f (k, s ₁ (k−1)) = s ₁ (k−1) (where s ₁ represents an input signal sequence to the chaos generator). , K = 0, 1,... The chaos coded modulation / demodulation method according to means 1, characterized in that:
According to the means 2, on the transmission side, an input signal sequence is obtained by performing an input side operation f which is an operation within the convergence range of the chaos generator, with the input side operation step taking the transmission information bit string and the fed back chaotic sequence as inputs. The chaotic sequence generation step generates the chaotic sequence by the chaotic generator with the input signal sequence as the input, and the output side arithmetic step performs the predetermined output side arithmetic with the chaotic sequence as the input to generate the chaotic transmission signal sequence. Generate. On the other hand, on the receiving side, the estimated transmission sequence generation step generates an estimated transmission sequence, and the input side calculation step performs the same input side calculation f as that of the transmission side with the estimated transmission sequence and the feedback estimated chaotic sequence as inputs. To generate an estimated input signal sequence, the estimated chaos sequence generation step receives the estimated input signal sequence as an input, generates an estimated chaos sequence by the same chaos generator as the transmission side, and the output side computation step transmits the estimated chaos sequence as an input The estimated transmission signal sequence is generated by performing the same output side calculation as the transmission side, and the decoding step calculates the error between the received signal sequence received from the transmission side and the estimated transmission signal sequence, and gives the estimated transmission sequence that gives the minimum error Is output as a decoding result of the received signal sequence.

3. 3. The chaotic coded modulation / demodulation method according to claim 2, wherein the predetermined output side operation h is expressed by the following mathematical formula .
However, b represents a digital transmission information bit string {0, 1}, where n, k = 0, 1,.
According to the means 3, on the transmission side, an input signal sequence is obtained by performing an input-side operation f that is an operation within the convergence range of the chaos generator, with the input-side operation step as an input of the transmission information bit string and the fed back chaotic sequence. The chaos sequence generation step generates a chaos sequence by the chaos generator with the input signal sequence as input, and the output side operation step generates the chaos transmission signal sequence by performing the output side operation h with the chaos sequence as input. To do. On the other hand, on the receiving side, the estimated transmission sequence generation step generates an estimated transmission sequence, and the input side calculation step performs the same input side calculation f as that of the transmission side with the estimated transmission sequence and the feedback estimated chaotic sequence as inputs. To generate an estimated input signal sequence, the estimated chaos sequence generation step receives the estimated input signal sequence as an input, generates an estimated chaos sequence by the same chaos generator as the transmission side, and the output side computation step transmits the estimated chaos sequence as an input The estimated transmission signal sequence is generated by performing the same output side calculation h as the transmission side, and the decoding step calculates the error between the received signal sequence received from the transmission side and the estimated transmission signal sequence, and the estimated transmission giving the minimum error The sequence is output as a decoding result of the received signal sequence. The output signal has a Euclidean distance of 3 or more due to the difference of b (n).

4). The chaotic coded modulation / demodulation method according to claim 3 , wherein s ₂ (k) is expressed by the following mathematical formula .
Oite the means _{4, s} 2 (k) is derived by generation equation annular chaos.

5. The chaos generator uses an annular chaos generation equation, the initial values are x ₀ = Re [s _i (k)], y ₀ = Im [s _i (k)], and are expressed by the following equations. chaotic coded modulation demodulating method according to any one of means 2 to 4, characterized in that the stuff.
s ₁ (k) = x ₂₀ + ji ₂₀
According to the means 5, the expression of xl + 1 and yl + 1 is repeated 20 times for each output from the expression of s ₁ (k). The output transmission signal is a chaotic shift keying (CSK) whose amplitude and phase change.

6). The chaotic coded modulation / demodulation method according to claim 2 , wherein the predetermined output-side operation h is expressed by the following equation .
However, b represents a digital transmission information bit string {0, 1}, where n, k = 0, 1,.
According to the means 6, on the transmission side, the input side operation step performs the input side operation f which is an operation within the convergence range of the chaos generator by using the transmission information bit string and the fed back chaotic sequence as inputs. The chaos sequence generation step generates a chaos sequence by the chaos generator with the input signal sequence as input, and the output side operation step generates the chaos transmission signal sequence by performing the output side operation h with the chaos sequence as input. To do. On the other hand, on the receiving side, the estimated transmission sequence generation step generates an estimated transmission sequence, and the input side calculation step performs the same input side calculation f as that of the transmission side with the estimated transmission sequence and the feedback estimated chaotic sequence as inputs. To generate an estimated input signal sequence, the estimated chaos sequence generation step receives the estimated input signal sequence as an input, generates an estimated chaos sequence by the same chaos generator as the transmission side, and the output side computation step transmits the estimated chaos sequence as an input The estimated transmission signal sequence is generated by performing the same output side calculation h as the transmission side, and the decoding step calculates the error between the received signal sequence received from the transmission side and the estimated transmission signal sequence, and the estimated transmission giving the minimum error The sequence is output as a decoding result of the received signal sequence. The modulation signal is a chaotic phase shift keying (CPSK) that varies only in phase with an amplitude of 1.

7). The chaotic coded modulation / demodulation method according to claim 2 , wherein the predetermined output-side operation h is expressed by the following equation .
However, b represents a digital transmission information bit string {0, 1}, where n, k = 0, 1,.
According to the means 7, on the transmission side, the input side operation step performs the input side operation f which is an operation within the convergence range of the chaos generator by using the transmission information bit string and the fed back chaotic sequence as inputs. The chaos sequence generation step generates a chaos sequence by the chaos generator with the input signal sequence as input, and the output side operation step generates the chaos transmission signal sequence by performing the output side operation h with the chaos sequence as input. To do. On the other hand, on the receiving side, the estimated transmission sequence generation step generates an estimated transmission sequence, and the input side calculation step performs the same input side calculation f as that of the transmission side with the estimated transmission sequence and the feedback estimated chaotic sequence as inputs. To generate an estimated input signal sequence, the estimated chaos sequence generation step receives the estimated input signal sequence as an input, generates an estimated chaos sequence by the same chaos generator as the transmission side, and the output side computation step transmits the estimated chaos sequence as an input The estimated transmission signal sequence is generated by performing the same output side calculation h as the transmission side, and the decoding step calculates the error between the received signal sequence received from the transmission side and the estimated transmission signal sequence, and the estimated transmission giving the minimum error The sequence is output as a decoding result of the received signal sequence. Note that the amplitude of the transmission signal is changed by the parity of the convolutional code, and the Euclidean distance is extended by the difference in the past b (n).

8). 8. The chaotic coded modulation / demodulation method according to claim 1 , wherein the output side operation generates a chaotic transmission signal sequence in which both amplitude and phase change in a pseudo-noise manner from the chaotic sequence .
According to the means 8, since the output side operation generates a chaotic transmission signal sequence in which both amplitude and phase change in a pseudo-noise manner from the chaotic sequence, it is possible to transmit information with extremely high confidentiality.

9. 8. The chaotic coded modulation and demodulation method according to claim 1, wherein the output side operation generates a chaotic transmission signal sequence in which only one of amplitude or phase changes from the chaotic sequence .
According to the means 9, since the output side operation generates a chaotic transmission signal sequence in which only one of the amplitude and phase changes from the chaotic sequence, the chaotic transmission signal sequence can be generated using an inexpensive device. .

10. The decoding step on the receiving side performs the error calculation by discarding the estimated transmission signal sequence by ½ each time the decoding constraint length is increased by 1 for each sequence having a predetermined length vl or more. 10. A chaotic code modulation demodulation method according to any one of means 1 to 9 .
According to the means 10, an error is calculated by performing a full search for a sequence of less than a predetermined length vl, and an estimated transmission signal sequence for each sequence of vl or more is incremented by 1 each time. Since the error calculation is performed by discarding ½ each, the number of calculation sequences can always be kept at 2 ^vl , and the decoding constraint length is increased while the divergence of the calculation amount is prevented, thereby reducing the decoding error rate. Can do.

11. A coded modulation demodulation method for transmitting information by coded modulation between a transmitting side and a receiving side,
The sender side
A transmission information bit sequence is input to a chaos generator to generate a coded signal sequence, and a chaos transmission signal sequence generation step for transmitting to the reception side as a chaos transmission signal sequence of a coded modulation scheme is provided,
The receiving side
An estimated transmission sequence generating step for generating an estimated transmission sequence;
An estimated transmission signal sequence generating step for generating an estimated transmission signal sequence by the same processing as the chaotic transmission signal sequence generating step on the transmission side with the estimated transmission sequence as an input;
A decoding step of calculating an error between the received signal sequence received from the transmitting side and the estimated transmission signal sequence, and outputting an estimated transmission sequence giving the minimum error as a decoding result of the received signal sequence;
With
The decoding step on the receiving side performs the error calculation by discarding the estimated transmission signal sequence by ½ each time the decoding constraint length is increased by 1 for each sequence having a predetermined length vl or more. A chaotic coded modulation and demodulation method characterized by the above.
According to the means 11, on the transmission side, the chaos transmission signal sequence generation step inputs the transmission information bit string to the chaos generator to generate the encoded signal sequence, and transmits it to the reception side as the chaotic transmission signal sequence of the coded modulation scheme. Then, on the receiving side, the estimated transmission sequence generation step generates an estimated transmission sequence, and the estimated transmission signal sequence generation step receives the estimated transmission sequence as an input and performs the same processing as the transmission side chaotic transmission signal sequence generation step. The decoding step calculates an error between the received signal sequence received from the transmitting side and the estimated transmission signal sequence, and outputs an estimated transmission sequence that gives the minimum error as a decoding result of the received signal sequence. Therefore, by performing analog coding using a chaotic sequence, that is, coding by a signal waveform, it is possible to achieve both good transmission characteristics and transmission confidentiality. That is, relatively good transmission error rate characteristics can be obtained even with low received signal quality, and sequential decoding is possible. Furthermore, since the communication signal sequence varies randomly due to chaos, it is close to noise, and it is possible to transmit information with excellent confidentiality that cannot be easily decoded by other people. In other words, because the transmission signal changes in a pseudo-noise manner by using a chaotic sequence, the transmission information is not clear in the signal itself, and all parameters in the chaotic transmission signal sequence generation step used on the transmission side and the reception side must be grasped. The secrecy of the transmission is realized because it is almost impossible for others to decode due to the uncorrelatedness of the chaotic sequence. Any type of chaos generator may be used, and a plurality of systems may be mixed. The degree of freedom of calculation in the encoding process is high, and the encoding rate can be set freely.
In particular, a full search is performed for a sequence of less than a predetermined length vl to perform error calculation, and for each sequence of vl or more, the estimated transmission signal sequence is reduced by 1/2 each time the decoding constraint length is increased by 1. Since the error calculation is performed by discarding each one, the number of calculation sequences can always be kept at 2 ^vl , and the decoding error rate can be reduced by increasing the decoding constraint length while preventing the amount of calculation from diverging.

12 12. The chaotic coded modulation / demodulation method according to any one of means 1 to 11, wherein the decoding step on the receiving side variably sets a decoding constraint length .
According to the means 12, since the decoding step on the receiving side variably sets the decoding constraint length, a trade-off between the calculation amount and the bit error rate is obtained by the length of the decoding constraint length, and decoding is performed only on the receiving side. The bit error rate can be controlled as desired. The decoding constraint length is the length of a signal sequence used to obtain a decoding result. In the present invention, it means the length of a received signal sequence or an estimated transmission signal sequence that is an error calculation target in a decoding step. .
13. 13. The chaotic coded modulation and demodulation method according to claim 12 , wherein the decoding step on the receiving side variably sets the decoding constraint length based on a measure of the probability of decoding bits .
According to the means 13, since the decoding step at the receiving side variably sets the decoding constraint length based on a measure of the probability of the decoded bit, when the probability of the decoded bit is high, the decoding constraint length is set short. Therefore, when the probability of decoding bits is low, the decoding constraint length can be set long to widen the search range and improve the certainty.
14 Said decoding step in the receiver side, compared the estimated transmission minimum error d between the received signal sequence when decoded bits in the sequence is set to 0 ₀ and a minimum error d ₁ in the case where the decoded bits and 1 When the minimum error d ₁ is smaller, the decoded bit is decoded as 1, and when the minimum error d ₀ is smaller, the decoded bit is decoded as 0. The chaos coding modulation demodulation method in any one of.
According to the means 14, when the minimum error d ₁ is smaller than the minimum error d _0, 1 is a more probable value than 0, so the decoded bit is decoded as 1 and the minimum error d ₁ When the minimum error d ₀ is smaller than 0, since 0 is more probable value than 1, the decoded bit is decoded as 0. If the minimum error d ₀ is equal to the minimum error d ₁ , the probability of the probability does not match between 0 and 1, so the decoded bit may be decoded to an arbitrary value (0 or 1).

15. The decoding step at the receiving side includes the minimum error d. ₀ And the minimum error d ₁ The absolute value of the difference between the difference is compared with a threshold value greater than or equal to 0. If the absolute value of the difference is greater than or equal to the threshold value, decoding is performed. ₀ , D ₁ The chaos coded modulation / demodulation method according to claim 14, wherein the chaos is recalculated.
In means 15, the minimum error d ₀ And the minimum error d ₁ If the absolute value of the difference between the received signal sequence and the estimated transmission signal sequence is equal to or greater than the threshold, the minimum error d between the received signal sequence and the estimated transmission signal sequence ₀ , D ₁ The difference between is sufficiently large, and a reliable decoding result can be obtained. Therefore, in this case, an increase in the amount of calculation can be suppressed by performing decoding without increasing the decoding constraint length. On the other hand, the minimum error d ₀ And the minimum error d ₁ If the absolute value of the difference between and is less than the threshold, the minimum error d ₀ , D ₁ The difference between the two is small, and a reliable decoding result cannot be obtained. Therefore, in this case, the decoding constraint length is increased and d ₀ , D ₁ And recalculate the minimum error d ₀ And the minimum error d ₁ By repeating the increase of the decoding constraint length and the error calculation until the absolute value of the difference between and becomes equal to or greater than the threshold value, a more reliable decoding result can be obtained.
16. The chaos transmission signal sequence generation step on the transmission side uses a plurality of chaos generators to generate a chaos transmission signal sequence using a chaos generator that varies depending on the value of the transmission information bit. The chaos coding modulation demodulation method in any one of.
According to the means 16, since the chaotic transmission signal sequence is generated by different chaos generators depending on whether the value of the transmission information bit is 0 or 1, the same chaotic transmission signal sequence is obtained unless the same 0, 1 sequence is given. Since it cannot be obtained (in other words, the decoding constraint length is infinite), a good transmission error rate characteristic can be obtained.
17. The chaos transmission signal sequence generating step on the transmission side generates the chaos transmission signal sequence using a plurality of chaos generators connected in cascade, parallel or a combination thereof. The chaos coding modulation demodulation method in any one of.
According to the means 17, the chaos transmission signal sequence generation step generates a chaos transmission signal sequence using a plurality of chaos generators connected in cascade or in parallel or a combination thereof. It is possible to increase the confidentiality of information transmission.

18. The chaotic transmission signal sequence generation step on the transmission side includes:
Any one of means 1 to 17, further comprising a packetizing step of packetizing the chaotic transmission signal sequence and temporarily terminating generation of the chaotic sequence by the chaotic generator to initialize the chaotic generator. Chaos coded modulation demodulation method.
  According to the means 18, the chaotic transmission signal sequence generating step packetizes the chaotic transmission signal sequence in the packetizing step, and once completes generation of the chaotic sequence by the chaotic generator and initializes the chaotic generator. Even when a decoding error occurs in a part of the bits, it is possible to suppress the error from being propagated to a sequence subsequent to the bit.
19. 19. The chaotic coded modulation and demodulation method according to claim 18, wherein the packetizing step inserts a tail bit when terminating each packet.
According to the means 19, since the packetizing step inserts a tail bit when terminating each packet, bit errors near the end of the packet can be suppressed.
20. 20. The chaos according to any one of means 1 to 19, further comprising a coding step of another scheme that is executed before or after the chaotic transmission signal sequence generation step and allows a modulation signal that changes in a noise manner. Coded modulation demodulation method.
According to the means 20, the amount of calculation in the decoding step can be further reduced by combining the chaotic transmission signal sequence generation step and the encoding step of another method. For example, as an encoding step of another method, other encoding (such as a turbo code) that allows a noise-changing modulation signal, a MIMO (Multiple-input multiple-output) transmission method, a multiplex transmission method, or the like is used. Can do.

  According to the present invention, it is possible to achieve both good transmission characteristics and transmission confidentiality by performing analog encoding using a chaotic sequence, that is, encoding by a signal waveform. That is, relatively good transmission error rate characteristics can be obtained even with low received signal quality, and sequential decoding is possible. Furthermore, since the communication signal sequence varies randomly due to chaos, it is close to noise, and it is possible to transmit information with excellent confidentiality that cannot be easily decoded by other people.

  DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments embodying a chaos coded modulation / demodulation method of the present invention will be described below in detail with reference to the drawings. First, the configuration of this embodiment will be described, and then the transmission characteristics will be described using simulation results.

  FIG. 1 shows a configuration of a chaos encoder provided on the transmission side. The chaos encoder inputs a transmission information bit string to a chaos generator, generates an encoded signal sequence, and transmits the encoded signal sequence to a reception side as a chaotic transmission signal sequence of an encoded modulation scheme (chaos transmission signal sequence generation step).

The specific contents of the chaotic transmission signal sequence generation step executed in the chaotic encoder will be described below. First, a digital transmission information bit string b (n) ε {0, 1}, (n = 0, 1,...) Is input to the arithmetic unit f bit by bit, and an input signal sequence s _i to the chaos generator is obtained. Is generated (input-side calculation step). At this time,
s _i (k) = f (k, b (n), s ₁ (k−1)), (k = 0, 1,...) (1)
It is represented by s _i (k−1) is a chaotic signal to which feedback to be described later is applied. f may be any calculation as long as it is within the convergence range of the chaos generation unit at the subsequent stage.

Next, using this s _i (k) as an input, the chaos generator generates a vector s ₁ (k) having chaos characteristics (chaos sequence generation step). This becomes an output signal and at the same time, becomes an input of the next s _i (k + 1), and continues the generation of the chaotic sequence. The initial value s ₁ (−1) is given in advance.

And s ₁ (k) is calculated by the output side calculation h.
s (k) = h (b (n), s ₁ (k)) (2)
As a result, the output signal s (k) is obtained (output side calculation step). This becomes a chaotic modulation signal (the chaotic transmission signal sequence of the present invention).

  In addition, like the modified example of the chaos encoder shown in FIG. 2, a plurality of chaos generators (chaos generator 1 and chaos generator 2) are provided, and the chaos generator is changed by an input bit or an input sequence in the operation f. It is also possible to configure so as to. Alternatively, s (k) may be used as an input to another chaos generator, and the chaos generators may be cascaded in multiple stages or connected in parallel to be multiplexed with other chaos sequence signals.

  As shown in FIG. 1, in the present embodiment, a transmission signal corresponding to each transmission bit string is generated by feeding back a chaotic sequence and performing an operation with transmission bits. If they are different, the signal sequence is completely different. That is, the transmission signal s (k) generated in the equations (1) and (2) is always b (0),. . , B (n-1) also varies depending on the value. Accordingly, when the present embodiment is viewed as a coded modulation system, it can be said that the code constraint length can be arbitrarily extended.

FIG. 3 shows a configuration of a chaos decoder provided on the receiving side. In the figure, the same chaos generators as those on the transmission side are used for the chaos sequence generators such as the chaos generator and the calculation units f and h. First, in the block shown in the left end of FIG. 3 in the chaos decoder, an estimated transmission sequence represented by the following equation is generated (estimated transmission sequence generation step).

Next, an estimated transmission signal sequence represented by the following formula is generated by the same processing as the transmission-side chaotic transmission signal sequence generation step using this estimated transmission sequence (the above formula ) as an input (estimated transmission signal sequence generation step).

  That is, the estimated transmission signal sequence generation step receives the estimated transmission sequence and the feedback estimated chaotic sequence as inputs, and performs the same input side operation f as the transmission side to generate an estimated input signal sequence, Estimated chaos sequence generation step by which the estimated chaos sequence is generated by the same chaos generator as the transmitting side with the estimated input signal sequence as an input, and the estimated transmission is performed by performing the same output side calculation h as the transmitting side with the estimated chaos sequence as an input And an output side calculation step for generating a signal sequence.

Next, an error Er between the received signal sequence r (k) received from the transmitting side and the estimated transmission signal sequence (the above formula) is calculated. The error Er is expressed as follows.

Then, the estimated transmission sequence giving the minimum error min | Er | is used as a decoding result (decoding step). Here, err () is a function for deriving some distance, for example, a square Euclidean distance represented by the following equation.

Decoding can be performed independently for each n, but it is also possible to perform lump decoding with l bits (l> 0) as one frame as shown below.

In this case, it is necessary to generate and compare all sequences of the estimated transmission sequence [ Equation 10 ]. When the square Euclidean distance is used as a function of error, Er is expressed as follows.

  In the equation (4), l corresponds to the decoding constraint length. Further, the frame length (decoding constraint length) may be adaptively changed by using a threshold value. A method for variably setting the decoding constraint length will be described below.

The estimated transmission sequence when decoded bit (first bit) is 0 in the formula, represents respectively one after another wherein the estimated transmission sequence when decoded bit is 1.

Further, the respective minimum errors min | Er | are represented as d ₀ and d ₁ and parameters shown in the equation (5) are introduced.
d [b (n)] = d ₀ −d ₁ (5)

When d [b (n)]> 0, b (n) = 1, and when d [b (n)] <0, when b (n) = 0 (d [b (n)] = 0 (Optional) That is, when the minimum error d ₁ is smaller than the minimum error d _0, 1 is a more probable value than 0, so that the decoded bit is decoded as 1, and the minimum error is smaller than the minimum error d _1. When d ₀ is smaller, 0 is a more probable value than 1, so the decoded bit is decoded as 0. Further, when the minimum error d ₀ and the minimum error d ₁ are equal, since 0 and 1 do not give the superiority or inferiority, the decoded bit is decoded to an arbitrary value (0 or 1).

FIG. 4 shows the concept of this decoding method. In the sequence of 0 and 1 in the lower part of FIG. 4, the leftmost bit enclosed by a square is a decoded bit, and the other bits are bits required to obtain a decoded result, and the length is l bits as a whole. ing. Using a certain threshold sh (≧ 0) for this d [b (n)], if | d [b (n)] | ≧ sh, decoding is performed, otherwise l → l + 1 is set to d ₀ , d ₁ Executes an algorithm that performs recalculation. That is, if the absolute value of the difference between the minimum error d ₀ and the minimum error d ₁ is not less than the threshold value, the difference between the minimum error d _0, d ₁ between the received signal sequence and the estimated transmission signal sequence is sufficiently large, probable A decoding result can be obtained. Therefore, in this case, an increase in the amount of calculation can be suppressed by performing decoding without increasing the decoding constraint length. On the other hand, when the absolute value of the difference between the minimum error d ₀ and the minimum error d ₁ is less than the threshold value, the difference between the minimum errors d ₀ and d ₁ is small and a reliable decoding result cannot be obtained. Therefore, in this case, the decoding constraint length is increased, d ₀ and d ₁ are recalculated, and the decoding constraint length is increased until the absolute value of the difference between the minimum error d ₀ and the minimum error d ₁ is equal to or greater than the threshold value. By repeating the error calculation, a more reliable decoding result can be obtained.

  As described above, the trade-off between the decoding calculation amount and the bit error rate (BER) can be controlled to some extent by setting the threshold sh. That is, if the threshold sh is increased, the calculation is often repeated until l becomes large, and if sh is too large, l satisfying | d [b (n)] | ≧ sh is increased and the amount of calculation is diverged. However, since the constraint length of the transmission code itself in the signal sequence s (k) is the same length as the signal sequence length, the probability that a decoding bit error will occur decreases as the decoding constraint length l increases. Become.

Therefore, this embodiment can realize a trade-off between the bit error rate and the calculation complexity only on the receiving side by setting the value of the constraint length l or sh even in the same received signal sequence. However, since the number of sequences of the estimated transmission sequence shown in the following equation when the decoding constraint length is 1 is 2 ^l , the amount of calculation immediately diverges as l increases.

Therefore, as shown in FIG. 5 at the time of calculation of equation (4), all the sequences up to 2 ^vl are searched, and after that, sequences with large Er in each region of b (n) = {0, 1} each time. Is discarded at a time, and the number of calculation sequences is always kept at 2 ^vl . This makes it possible to increase l while preventing divergence of the calculation amount. If v1 is small, the probability of erroneously discarding a correct decoding sequence increases, so even if l is increased, the effect is small. However, by increasing v1 and l, it is possible to reduce the decoding error rate only on the receiving side. . Also, once the estimated transmission sequence [Formula 1] is decoded in error, the subsequent chaotic sequence in the decoder does not match the transmission side, and the decoding error propagates to the subsequent bits and normal decoding cannot be performed. In order to prevent this error propagation, the chaotic transmission signal sequence may be packetized.

For example, packetization as shown in FIG. 6 is performed on the transmitter side, tail bits are inserted at the end of the packet, and a chaos generator is initialized at the end of the packet (packetization step). The initialization of the chaos generator is to return the chaos generator to the initial value s ₁ (−1). Similarly, by performing initialization at the end of the packet on the receiving side, it is possible to prevent error propagation to a subsequent packet even if an error occurs in the packet.

  In principle, it is only necessary to initialize the state of the chaos generator at the end of the packet without inserting tail bits, but in this case, the error probability increases because the data near the end of the packet cannot be extended. To do.

In the examination so far, the input b (n) is 1 bit. However, as shown in FIG. 7, it is also possible to input it to the operation f as b (n) of several bits. Any number of bits b (n) to be input to this once and _{n d,} the number of b s which is outputted to (n) (k) when a _{r c pieces,} with r c _{/ n} d> 1 For example, encoding to add redundancy is performed. At this time, when the packet configuration of FIG. 6 is included, the overall transmission efficiency is expressed by the following equation .

Consider transmission of a code that generates a transmission signal by using an encoder that operates only the output-side operation h in the subsequent stage using input bits as shown in FIG. The receiver also has a chaos generator having the same configuration, f, h, and an initial value s ₁ (−1).

A simulation was performed in an equivalent low-frequency transmission system of an equivalent low-frequency system and a Gaussian noise communication channel as shown in FIG. In the following, it is assumed that the receiver side is completely synchronized. The parameters of the transmission system are as shown in Table 1. The parameters in the encoder are the input side operation f,
f (k, s ₁ (k−1)) = s ₁ (k−1) (6)
And the output side calculation h is as follows.

Here, s ₂ (k) is as follows.

In the chaos generator, the initial value is set to x ₀ = Re [s _i (k)], y ₀ = Im [s _i (k)] using the generation equation of the annular chaos, and the following equation (9) , (10).
s ₁ (k) = x ₂₀ + ji ₂₀ (10)

From the expression (7), the output signal has a Euclidean distance of 3 or more due to the difference of b (n). Expression (8) is an expression derived from the chaos generation expression of Expression (9). From Expression (10), Expression (9) is repeated 20 times for each output. The initial value s ₁ (−1) was generated by a random number. The output transmission signal is a chaotic shift keying (CSK) whose amplitude and phase change. Further, when the output side calculation h is expressed by the equation (11), the modulation signal becomes a chaotic phase shift keying (CPSK) in which only the phase varies with an amplitude of 1.

Here, a transmission signal sequence configured by CSK is shown in FIG. 10 on the IQ plane in the baseband. This point sequence is a characteristic of the chaotic sequence, and shows an uncorrelated transition that differs depending on the initial value s ₁ (−1), the transmission information bit sequence, the type of the chaotic equation, etc. It is difficult to decode the original data sequence b (n) except when frame synchronization and symbol synchronization are acquired and b (n) before that time is correctly grasped. Therefore, it can be said that this method is a method having both high confidentiality. From Table 1, the overall transmission efficiency is about 0.1 bit / symbol.

FIG. 11 shows the configuration of the decoder. The chaos generation part in the figure is the same as that of the encoder of FIG. For the calculation of the error, a square Euclidean distance of equation (4) was used, a threshold sh was set for the parameter of equation (5), the constraint length was variable, and a decoding method of decoding bit by bit was applied. Here, as shown in Table 1, v _l = 8 to 13 and the maximum constraint length was l _max .

That is, in decoding, first, v _l = 8 is set, and decoding calculation is performed according to the equations (4) and (5). If d [b (n)] in equation (5) exceeds sh by the time when the constraint length reaches l _max , a determination is made and the process proceeds to the next bit. When l _max is reached without exceeding sh, the calculation is repeated in the same manner as v _l → v _l +1. If v _l = 13 and the constraint length becomes l _max but does not exceed sh, the determination is made based on the value of b (n) at which | d [b (n)] | Do.

In the simulation, the average decoding constraint length and the average v _l (average decoding state index) when Eb / N0 was 2 to 10 dB and 5 frames (10 ⁵ bits) were transmitted were evaluated. The calculation results are shown in FIG. In this setting, no decoding error occurred in any case. As shown in the figure, as Eb / N0 increases, the average decoding constraint length decreases and the average v _l also decreases. With the improvement of Eb / N0, v _l is slightly lower than 8. This is because the decoding complexity is effectively reduced to the equivalent of v _l = 1 in the vicinity of the tail bit.

As described above, performs encoding based modulation chaotic equations, by setting adaptively vary the decoding constraint length using a threshold, error-free transmission at Eb / N0 = 2 dB is realized in the transmission of 10 ⁵ bits It was. By increasing the values of v _l and l _max , it can be expected that this characteristic is further improved in a trade-off relationship with the calculation amount. Furthermore, the transmission sequence has a feature that it cannot be easily decoded by a third party. It is considered that various transmission characteristics can be realized depending on the chaotic sequence used, coding rate, frame length, tail bit length, and decoding constraint length.

  Next, as shown in FIG. 13, a trellis encoder is connected to the previous stage of the chaos encoder, and the characteristics when b (n) is the parity of the trellis code were evaluated. As shown in this example, the present apparatus can be used in combination with other transmission methods.

The trellis code used is a recursive systematic convolutional code RSC [15/7], b (n) is a parity bit of the trellis codeword, and no uncoded information bits are transmitted. Therefore, there is no decrease in transmission efficiency due to the trellis encoder, and the overall transmission efficiency is the same as in the first embodiment. The parameters in the encoder are as shown in the following equation (12) in the part (7) of the first embodiment. The amplitude of the transmission signal is changed by the parity of the convolutional code, and the difference in the past b (n) The Euclidean distance was also set to be longer. The rest is the same as the first embodiment.

On the decoding side, since the calculations of equations (4) and (5) with v _l = 8 to 13 include the trellis diagram of the convolutional code, the decoding is performed in the same manner as in the first embodiment.

Simulation conditions as well as Eb / N0 of Example 1 2~10dB, _{v l} (average decoding condition index) and average 5 frames ^(105-bit) average decoding constraint length at the time of transmission was calculated. The results are shown in FIG. Also in this example, no decoding error occurred in either case. As in FIG. 12, as Eb / N0 increases, the average decoding constraint length decreases, the average v _l also decreases, and both are further reduced compared to the case of no encoding. That is, it is shown that the decoding calculation amount is reduced by combining with an external encoder. As described above, it is considered that various transmission forms can be realized by combining with an external transmission system using chaotic coded modulation.

It is a block diagram which shows the structure of a chaos encoder. It is a block diagram which shows the modification of the encoder which used two chaotic generators. It is a block diagram which shows the structure of a chaos decoder. It is a conceptual diagram of the decoding using a frame. It is the figure explaining the outline | summary about the reduction method of the number of exploration series. It is a figure which shows the structural example of the packet which consists of a data bit and a tail bit. It is a figure which shows the example of an input bit number and the calculation f. 3 is a block diagram illustrating a configuration of an encoder used in Embodiment 1. FIG. It is a block diagram of the transmission system using the chaos code used for simulation. An example of a transmission signal sequence of CSK is shown. It is a block diagram which shows the structure of the decoder used in Example 2. FIG. It is a graph which shows the average decoding constraint length of decoding at the time of transmission of a chaos coding modulation demodulation method, and an average decoding state index. It is a block diagram which shows the structure of the chaotic encoder which connected the convolutional code. It is a graph which shows the average decoding constraint length and average decoding state index | exponent of decoding at the time of transmission at the time of concatenating a convolutional code.

Claims (20)

A coded modulation demodulation method for transmitting information by coded modulation between a transmitting side and a receiving side,
The sender side
A transmission information bit sequence is input to a chaos generator to generate a coded signal sequence, and a chaos transmission signal sequence generation step for transmitting to the reception side as a chaos transmission signal sequence of a coded modulation scheme is provided,
The receiving side
An estimated transmission sequence generating step for generating an estimated transmission sequence;
An estimated transmission signal sequence generating step for generating an estimated transmission signal sequence by the same processing as the chaotic transmission signal sequence generating step on the transmission side with the estimated transmission sequence as an input;
A decoding step of calculating an error between the received signal sequence received from the transmitting side and the estimated transmission signal sequence, and outputting an estimated transmission sequence that gives the minimum error as a decoding result of the received signal sequence , and
The chaotic transmission signal sequence generation step on the transmission side includes:
An input-side operation step of generating an input signal sequence by performing a predetermined input-side operation with the transmission information bit string and the fed back chaotic sequence as inputs; and
A chaos sequence generation step of generating a chaos sequence by the chaos generator with the input signal sequence as an input; and
An output side computation step for generating the chaotic transmission signal sequence by performing a predetermined output side computation with the chaos sequence as an input; and
With
The estimated transmission signal sequence generation step on the receiving side includes:
An input side calculation step of generating an estimated input signal sequence by performing the same input side calculation as the transmission side with the estimated transmission sequence and the feedback estimated chaotic sequence as inputs,
An estimated chaos sequence generating step for generating an estimated chaos sequence by the same chaos generator as the transmission side with the estimated input signal sequence as an input;
A chaotic coded modulation / demodulation method comprising: an output-side operation step of generating an estimated transmission signal sequence by performing the same output-side operation as that of the transmitting side with the estimated chaotic sequence as an input . The predetermined input-side operation f is expressed by the formula f (k, s ₁ (k−1)) = s ₁ (k−1) (where s ₁ represents an input signal sequence to the chaos generator). , K = 0, 1,...) , The chaotic coded modulation and demodulation method according to claim 1. 3. The chaotic coded modulation / demodulation method according to claim 2 , wherein the predetermined output-side operation h is expressed by the following equation .
However, b represents a digital transmission information bit string {0, 1}, where n, k = 0, 1,. The chaotic coded modulation / demodulation method according to claim 3 , wherein the s ₂ (k) is expressed by the following mathematical formula .
The chaos generator uses an annular chaos generation equation, the initial values are x ₀ = Re [s _i (k)], y ₀ = Im [s _i (k)], and are expressed by the following equations. chaotic coded modulation demodulating method according to any one of claims 2 to 4, characterized in that a thing.
s ₁ (k) = x ₂₀ + ji ₂₀ 3. The chaotic coded modulation / demodulation method according to claim 2 , wherein the predetermined output-side operation h is expressed by the following equation .
However, b represents a digital transmission information bit string {0, 1}, where n, k = 0, 1,. 3. The chaotic coded modulation / demodulation method according to claim 2 , wherein the predetermined output-side operation h is expressed by the following equation .
However, b represents a digital transmission information bit string {0, 1}, where n, k = 0, 1,. 8. The chaotic coded modulation and demodulation method according to claim 1, wherein the output side operation generates a chaotic transmission signal sequence whose amplitude and phase change in a pseudo-noise manner from the chaotic sequence . 8. The chaotic coded modulation / demodulation method according to claim 1, wherein the output side operation generates a chaotic transmission signal sequence in which only one of amplitude and phase changes from the chaotic sequence. . The decoding step on the receiving side performs the error calculation by discarding the estimated transmission signal sequence by ½ each time the decoding constraint length is increased by 1 for each sequence having a predetermined length vl or more. A chaotic code modulation demodulation method according to any one of claims 1 to 9 . A coded modulation demodulation method for transmitting information by coded modulation between a transmitting side and a receiving side,
The sender side
A transmission information bit sequence is input to a chaos generator to generate a coded signal sequence, and a chaos transmission signal sequence generation step for transmitting to the reception side as a chaos transmission signal sequence of a coded modulation scheme is provided,
The receiving side
An estimated transmission sequence generating step for generating an estimated transmission sequence;
An estimated transmission signal sequence generating step for generating an estimated transmission signal sequence by the same processing as the chaotic transmission signal sequence generating step on the transmission side with the estimated transmission sequence as an input;
A decoding step of calculating an error between the received signal sequence received from the transmitting side and the estimated transmission signal sequence, and outputting an estimated transmission sequence giving the minimum error as a decoding result of the received signal sequence;
With
The decoding step on the receiving side performs the error calculation by discarding the estimated transmission signal sequence by ½ each time the decoding constraint length is increased by 1 for each sequence having a predetermined length vl or more. A chaotic coded modulation and demodulation method characterized by the above. 12. The chaotic coding modulation demodulation method according to claim 1, wherein the decoding step on the receiving side variably sets a decoding constraint length . 13. The chaotic coding modulation demodulation method according to claim 12 , wherein the decoding step on the receiving side variably sets the decoding constraint length based on a measure of the probability of decoding bits . Said decoding step in the receiver side, compared the estimated transmission minimum error d between the received signal sequence when decoded bits in the sequence is set to 0 ₀ and a minimum error d ₁ in the case where the decoded bits and 1 The decoded bit is decoded as 1 when the minimum error d ₁ is smaller, and the decoded bit is decoded as _{0 when} the minimum error d ₀ is smaller. 14. The chaos coding modulation demodulation method according to any one of the above. The decoding step at the receiving side includes the minimum error d. ₀ And the minimum error d ₁ The absolute value of the difference between the difference is compared with a threshold value greater than or equal to 0. If the absolute value of the difference is greater than or equal to the threshold value, decoding is performed. ₀ , D ₁ The chaos coded modulation / demodulation method according to claim 14, wherein the chaos is recalculated. The chaotic transmission signal sequence generation step on the transmission side uses a plurality of chaos generators to generate a chaos transmission signal sequence using a chaos generator that varies depending on a value of the transmission information bit. 15. The chaos coding modulation demodulation method according to any one of 15. 2. The chaotic transmission signal sequence generation step on the transmission side generates the chaotic transmission signal sequence using a plurality of chaos generators connected in cascade, parallel, or a combination thereof. 16. The method for demodulating chaotically coded modulation according to any one of 16 above. The chaotic transmission signal sequence generation step on the transmission side includes:
    The packetizing step of packetizing the chaotic transmission signal sequence and temporarily terminating generation of the chaotic sequence by the chaotic generator to initialize the chaotic generator. And chaotic coded modulation demodulation method. 19. The chaotic coded modulation and demodulation method according to claim 18, wherein in the packetizing step, a tail bit is inserted when terminating each packet. The encoding method according to any one of claims 1 to 19, further comprising an encoding step of another method that is executed before or after the chaotic transmission signal sequence generation step and allows a modulation signal that changes in a noise manner. Chaos coded modulation demodulation method.