Power local communication method based on Viterbi soft information demodulation

Abstract

The invention discloses an electric power local communication method based on Viterbi soft information demodulation, which comprises the following steps of: firstly, carrying out convolutional coding on information bits by a transmitting end, and modulating the information bits to an electric power authorization 230MHz frequency band by adopting a CPFSK (Continuous Phase Frequency Shift Keying) mode; then up-converting the information bits, sending the information bits to a receiving end through an electric power authorization frequency band, and down-converting the information bits by the receiving end; then, a receiving end extracts soft information in the CPFSK signal by adopting a soft information calculation method based on a Viterbi demodulation algorithm; and finally, a decoder performs soft decoding based on soft information by adopting a convolutional code decoding technology and utilizing a log-likelihood ratio. According to the invention, the reliability and communication rate of the local communication system are improved.

Description

Technical Field

[0001] The invention relates to the technical field of wireless communication, in particular to a local power communication method based on Viterbi soft information demodulation. Background Art

[0002] Continuous Phase Frequency Shift Keying (CPFSK) is a continuous phase frequency modulation method that changes the frequency of the signal so that it can be demodulated at the receiving end. High spectrum utilization and high power utilization are its two most significant features. It is very suitable for application scenarios with high information rate requirements, so it has become a research object that has attracted much attention in the field of wireless communication technology.

[0003] The demodulation of CPFSK signals is divided into two categories: coherent demodulation and incoherent demodulation. The commonly used schemes for incoherent demodulation in the early stage are the limiter-discriminator method (LD method) and the differential detection method. The LD method and the differential detection method have simple structures, and because they do not utilize the intrinsic correlation of the signal phase, their demodulation performance is far inferior to coherent demodulation when the signal-noise ratio is small. In 1999, Colavolpe et al. proposed a noncoherent sequence detection algorithm (NSE algorithm), which makes the most of the memory characteristics of the CPFSK signal, so it has significantly improved the performance compared with the previous algorithms. However, in order to simplify the complexity, the NSE algorithm adopts some approximate processing, which can only be called a quasi-optimal maximum likelihood detection algorithm. Under the AWGN channel, the best demodulation method for CPFSK signals is coherent maximum likelihood sequence detection, which is widely used in the demodulation implementation of continuous phase modulation signals.

[0004] At the same time, in modern communication systems, CPFSK modulation can be combined with channel coding technology to obtain better system anti-noise and anti-interference performance. Among them, Turbo code and LDPC code coding technology have been widely used in recent years due to their superior performance and moderate complexity. In order to ensure the performance of its decoding, the input of the decoding end must be soft decision information, so the soft decision demodulation of CPFSK signals becomes a core problem that needs to be solved.

[0005] The traditional CPFSK-based communication system adopts a demodulation scheme based on hard decision, which cannot make full use of channel information. Therefore, its transmission rate is relatively low and cannot meet the requirements of high-speed transmission. Summary of the invention

[0006] The object of the present invention is to provide a local power communication solution with high reliability and transmission rate.

[0007] The technical solution to achieve the purpose of the present invention is: a power local communication method based on Viterbi soft information demodulation, comprising the following steps:

[0008] Step 1: The transmitter performs convolution encoding on the information bits;

[0009] Step 2: Use phase-continuous constant envelope modulation CPFSK to modulate the information bits into the power authorized 230MHz frequency band;

[0010] Step 3: up-convert the information bits and send them to the receiving end through the power authorization frequency band, and the receiving end down-converts the information bits;

[0011] Step 4: The receiving end uses a soft information calculation method based on the Viterbi demodulation algorithm to extract soft information from the CPFSK signal;

[0012] Step 5: Decode the soft information to obtain information bits.

[0013] Furthermore, in step 2, the information bits are modulated to the power authorized 230 MHz frequency band, and the signal bandwidth after CPFSK modulation is less than 25 kHz, and the signal operates in the power 230 MHz authorized frequency band.

[0014] Furthermore, the receiving end described in step 4 adopts a soft information calculation method based on the Viterbi demodulation algorithm to extract the soft information in the CPFSK signal, as follows:

[0015] Assume that the frequency corresponding to the information symbol "0" is f0, the transmitted signal is s0(t), the frequency corresponding to the information symbol "1" is f1, the transmitted signal is s1(t), and the receiving end uses the Viterbi decoding algorithm to calculate the branch metric of each bit. The likelihood function of the transmitted signal s0(t) and the transmitted signal s1(t) is:

[0016]

[0017] Where r(t) represents the received signal, σ n 2 represents the noise variance;

[0018] The log-likelihood ratio (soft decision information) of information symbol "0" and information symbol "1" is:

[0019]

[0020]

[0021] Furthermore, the decoding of the soft information in step 5 is specifically that the decoder adopts a convolutional code decoding technology and performs soft decoding based on the soft information using a log-likelihood ratio.

[0022] Compared with the prior art, the present invention has the following significant advantages: (1) it has excellent anti-interference capability, and through coding it can further enhance the system's resistance to channel transmission errors, thereby increasing the transmission rate of the communication system; (2) the result is obtained through Viterbi decoding calculation, which reduces the bit error rate and improves the reliability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a flow chart of a method for local power communication of the present invention.

[0024] Figure 2 4 is a demodulation performance curve diagram of the power local communication method in an embodiment of the present invention. DETAILED DESCRIPTION

[0025] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0026] Combination Figure 1 The present invention provides a power local communication method based on Viterbi soft information demodulation, comprising the following steps:

[0027] Step 1: The transmitter performs convolution encoding on the information bits;

[0028] Step 2: Use phase-continuous constant envelope modulation CPFSK to modulate the information bits to the power authorized 230MHz frequency band. The signal bandwidth after CPFSK modulation is less than 25kHz, and it works on the power 230MHz authorized frequency band.

[0029] Step 3: up-convert the information bits and send them to the receiving end through the power authorization frequency band, and the receiving end down-converts the information bits;

[0030] Step 4: The receiving end uses a soft information calculation method based on the Viterbi demodulation algorithm to extract the soft information in the CPFSK signal, as follows:

[0031] Assume that the frequency corresponding to the information symbol "0" is f0, the transmitted signal is s0(t), the frequency corresponding to the information symbol "1" is f1, the transmitted signal is s1(t), and the receiving end uses the Viterbi decoding algorithm to calculate the branch metric of each bit. The likelihood function of the transmitted signal s0(t) and the transmitted signal s1(t) is:

[0032]

[0033] Where r(t) represents the received signal, σ n2 represents the noise variance;

[0034] The log-likelihood ratio (soft decision information) of information symbol "0" and information symbol "1" is:

[0035]

[0036] Step 5: The decoder uses convolutional code decoding technology and log-likelihood ratio to perform soft decoding based on soft information, as follows:

[0037] The M-ary CPFSK signal can be expressed as:

[0038]

[0039] Among them, f c is the carrier frequency, f d is the peak frequency deviation (I n = ±1 when the frequency deviation of the carrier), φ0 is the initial phase;

[0040]

[0041] Wherein, φ(t;I) is the additional phase of s(t) minus the carrier phase;

[0042] Although the amplitude modulated PAM signal expression d(t) is discontinuous, its integral is continuous, so φ(t; I) is a continuous phase signal in the interval nT≤t≤(n+1)T. Therefore, in the time interval [nT,(n+1)T], the additional phase is:

[0043]

[0044] In the formula,

[0045] h=2f d T

[0046]

[0047]

[0048] The parameter h is called the modulation index, θ n It represents the accumulated phase at time nT, that is, the memory value of all information symbols before nT.

[0049] Accumulated phase θ nAll possible values ​​of can be represented by a finite discrete set, called phase state, and the phase state of the CPFSK signal can be represented as a state grid diagram. It can be deduced that for a CPFSK signal with a modulation index of h, when h is a rational number, it can be represented as h = m / p, m and p are mutually prime positive integers, and the number of phase states is related to the parity of m.

[0050] When m is an even number, θ n There are p phase states:

[0051]

[0052] When m is an odd number, θ n There are 2p phase states:

[0053]

[0054] Assume that a CPFSK signal passes through an AWGN channel. Ignoring the frequency offset, the received signal is expressed as r(t)=s(t)+n(t), where s(t) is the transmitted CPFSK signal and n(t) is Gaussian white noise. At the transmitting end, different information symbols I correspond to different transmitted signals s m (t),m=1,2,…,M.

[0055] Under the AWGN channel, the likelihood function (i.e., conditional probability density function P(r(t) / s(t)))f(r(t) / s(t)) is:

[0056]

[0057] Since the likelihood function is a monotonic function, we take the logarithm of the likelihood function, because is a constant, the maximum value of the likelihood function is equivalent to The minimum value of , which is expanded to:

[0058]

[0059] in, is a common value, which has no effect on the minimum value, and the CPFSK signal has a constant envelope characteristic, so The size of is the same, and it has no effect on finding the minimum value, so find The minimum value of is equivalent to finding The maximum value of .

[0060] because

[0061]

[0062] Therefore, when demodulating according to the maximum likelihood decision criterion, the correlation metric can be As a branch metric for the Viterbi algorithm.

[0063] In summary, the branch metrics and path metrics in the Viterbi demodulation algorithm are derived as follows:

[0064] Assuming that the receiver in the CPFSK system performs down-conversion processing on the received signal, the baseband signal s(t) is:

[0065]

[0066] Where I is the information symbol sequence representing the modulation information, I n is the information symbol in [nT, (n+1)T], I n ∈I={±1,±3,±5,···,±(M-1)}.

[0067] For the transmitted signal S = {s1, s2, s3, ..., s M}, M information symbols I correspond to M transmitted signals S, and also correspond to M branch metrics. At the same time, the accumulated phase θ is established according to the modulation index h n The L phase states correspond to L path metrics.

[0068] At time t = nT, define the branch metric as Z n (θ n ;I n ), the path metric is CM n (θ n ;I n ), the path metric and branch metric are expressed as:

[0069]

[0070] Among them, CM n-1 (θ n-1 ;I n-1 ) represents the surviving path metric at time t = (n-1)T, and the latter term Represents the branch metric within the time interval (n-1)T≤t≤nT.

[0071] Then the soft-decision based Viterbi demodulation is derived as follows:

[0072] Taking the 2CPFSK signal as an example, the frequency corresponding to the information symbol "0" is set to f0, and the transmitted signal is s0(t); the frequency corresponding to the information symbol "1" is set to f1, and the transmitted signal is s1(t). Then the likelihood function of the transmitted signal s0(t) and the transmitted signal s1(t) is:

[0073]

[0074] The soft output Viterbi algorithm proposed in the present invention uses the log-likelihood ratio as soft information. The log-likelihood ratio of the information symbol "1" and the information symbol "0" can be derived from the likelihood function expression, that is, the soft decision information is:

[0075]

[0076] It can be seen that the log-likelihood ratio of information symbol "1" and information symbol "0" is finally simplified to the difference between the correlation values ​​of the received signal and the two transmitted signals, and these two correlation values ​​are related to the branch metric in the Viterbi algorithm. Therefore, by linking this log-likelihood ratio with the metric value in the Viterbi algorithm, the soft decision information LLR can be rewritten as:

[0077] LLR=[CM n-1 (θ n-1 ;I (n -1 )1 )+Z n (θ n ;I n1 )]-[CM n-1 (θ n-1 ;I (n -1 )0 )+Z n (θ n ;I n0 )]

[0078] In summary, the soft decision information output by the demodulator, that is, the log-likelihood ratio of "1" to "0", is the sum of the branch metric corresponding to the information symbol "1" and the surviving path metric corresponding to the previous moment of "1" minus the sum of the branch metric corresponding to the information symbol "0" and the surviving path metric corresponding to the previous moment of "0".

[0079] Example 1

[0080] This embodiment adopts a method for local power communication based on Viterbi soft information demodulation. First, the transmitter performs convolution encoding on the information bits, and uses phase-continuous constant envelope modulation, namely continuous phase frequency shift keying CPFSK, to modulate the information bits to the power-authorized 230MHz frequency band; then the information bits are up-converted and sent to the receiving end through the power-authorized frequency band, and the receiving end down-converts the information bits; then the receiving end uses a soft information calculation method based on the Viterbi demodulation algorithm to extract the soft information in the CPFSK signal; finally, the decoder uses convolutional code decoding technology and uses the log-likelihood ratio to perform soft decoding based on soft information. The bit error probabilities of the method of the present invention and those based on Viterbi hard information demodulation are plotted into curves for comparison. Figure 2 The performance of the soft information demodulation scheme based on Viterbi proposed in the present invention under 2FSK modulation is given, and it can be found that the demodulation performance is improved by about 3dB.

[0081] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for local power communication based on Viterbi soft information demodulation, characterized in that: The following steps are involved: Step 1: The transmitter performs convolution encoding on the information bits; Step 2: Use phase-continuous constant envelope modulation CPFSK to modulate the information bits into the power authorized 230MHz frequency band; Step 3: up-convert the information bits and send them to the receiving end through the power authorization frequency band, and the receiving end down-converts the information bits; Step 4: The receiving end uses a soft information calculation method based on the Viterbi demodulation algorithm to extract soft information from the CPFSK signal; Step 5: Decode the soft information to obtain information bits.

2. The power local communication method based on Viterbi soft information demodulation according to claim 1, characterized in that: The information bits described in step 2 are modulated to the power authorized 230 MHz frequency band, and the signal bandwidth after CPFSK modulation is less than 25 kHz, and it works on the power 230 MHz authorized frequency band.

3. The power local communication method based on Viterbi soft information demodulation according to claim 1, characterized in that: The receiving end described in step 4 uses a soft information calculation method based on the Viterbi demodulation algorithm to extract soft information from the CPFSK signal, as follows: Assume that the frequency corresponding to the information symbol "0" is f0, the transmitted signal is s0(t), the frequency corresponding to the information symbol "1" is f1, the transmitted signal is s1(t), and the receiving end uses the Viterbi decoding algorithm to calculate the branch metric of each bit. The likelihood function of the transmitted signal s0(t) and the transmitted signal s1(t) is: Where r(t) represents the received signal, σ n 2 represents the noise variance; The log-likelihood ratio of information symbol "0" and information symbol "1", i.e., the soft decision information, is:

4. The power local communication method based on Viterbi soft information demodulation according to claim 1, characterized in that: The decoding of the soft information in step 5 is specifically that the decoder adopts a convolutional code decoding technology and uses a log-likelihood ratio to perform soft decoding based on the soft information.

Images