A DSSS enhanced synchronization method combining PMF-FFT and improved Costas loop

Abstract

The present invention discloses a DSSS enhanced synchronization method combining PMF-FFT and an improved Costas loop, which is applied to the field of wireless communication technology. Traditional receiver designs typically employ independent frequency offset estimation and carrier synchronization methods, which are unsuitable for processing high-dynamic environments and low signal-to-noise ratio conditions. The present invention employs a maximum frequency offset estimation algorithm based on a partially matched filter-fast Fourier transform. By performing an FFT transform on the received signal, the frequency offset of the signal is accurately estimated, thereby effectively reducing the impact of the frequency offset on synchronization performance. Based on the frequency offset estimation, the receiver system designed in the present invention incorporates a Costas loop for carrier synchronization tracking. High-precision carrier recovery is achieved by locking the phase of the input signal.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless communications, and in particular relates to a radio signal carrier synchronization technology. Background Art

[0002] In modern wireless communication systems, receiver synchronization is crucial. Synchronization performance directly impacts the system's bit error rate, transmission rate, and overall communication quality. Traditional receiver designs typically employ separate frequency offset estimation and carrier synchronization methods. However, these methods still have limitations in handling highly dynamic environments and low signal-to-noise ratios. Separating frequency offset estimation and carrier synchronization not only increases system complexity but can also negatively impact carrier synchronization due to frequency offset estimation errors. Summary of the Invention

[0003] To overcome these challenges, the present invention designs a fast synchronization receiver system based on maximum frequency offset estimation, which achieves more efficient carrier synchronization by combining advanced algorithms and feedback mechanisms.

[0004] The technical solution adopted by the present invention is: a DSSS (Direct Sequence Spread Spectrum) enhanced synchronization method combining PMF-FFT and an improved Costas loop, wherein the frequency offset estimation process is implemented using a PMF-FFT algorithm module, which includes: a bandpass filter, a squarer, a pulse shaping filter, a PMF operation unit, and an FFT operation unit connected in sequence; after the received IQ signal passes through the PMF-FFT algorithm module, a coarse estimate of the maximum frequency offset is obtained; and a low-pass filter is also included, wherein the input of the low-pass filter is the output of the bandpass filter;

[0005] The carrier synchronization process is implemented using an improved Costas loop based on a PMF-FFT loop. The improved Costas loop includes a despreading module, a phase detector, a loop filter, and a digitally controlled oscillator. The despreading module inputs the output of the low-pass filter in the PMF-FFT algorithm module and the output of the digitally controlled oscillator. The despreading module output serves as the input of the phase detector, which in turn serves as the input of the loop filter. The loop filter output, along with a coarse estimate of the maximum frequency offset, serves as the input of the digitally controlled oscillator.

[0006] When the phase detector error is 0, locking is performed and a synchronized demodulated output is obtained.

[0007] Beneficial Effects of the Present Invention: The receiver system proposed in this invention utilizes a maximum frequency offset estimation algorithm based on a partial matching filter (PMF)-Fast Fourier Transformation (FFT). This algorithm accurately estimates the signal's frequency offset by performing an FFT transform on the received signal, effectively reducing the impact of frequency offset on synchronization performance. Compared to traditional frequency offset estimation algorithms, the PMF-FFT algorithm offers higher accuracy and faster convergence, providing more reliable frequency offset estimation in dynamically changing channel environments.

[0008] On the basis of frequency offset estimation, the receiver system designed by the present invention combines the Costas loop for carrier synchronization tracking. The Costas loop is a carrier synchronization technology widely used in digital communication systems. It achieves high-precision carrier recovery by locking the phase of the input signal. However, in a low signal-to-noise ratio environment, the performance of the traditional Costas loop will be significantly reduced. In order to solve this problem, the present invention introduces the despread output in the direct sequence spread spectrum system as a feedback mechanism in the Costas loop. Through this feedback mechanism, the system can effectively improve the synchronization performance of the Costas loop in a low signal-to-noise ratio environment and significantly reduce the bit error rate;

[0009] Compared to traditional receiver synchronization schemes, the system designed in this invention demonstrates significant advantages in both frequency offset estimation and carrier synchronization. By combining frequency offset estimation and carrier synchronization, the present invention leverages the high precision of the frequency offset estimation algorithm and the enhanced effect of the feedback mechanism to achieve a tighter synchronization process. Furthermore, the present invention's design maintains excellent synchronization performance in low signal-to-noise ratio and high-dynamic environments, demonstrating strong robustness and adaptability.

[0010] The method of the present invention has the following advantages:

[0011] 1. The present invention adopts PMF-FFT combined with Costas loop carrier synchronization method, which greatly improves the speed of carrier synchronization. When the received information signal-to-noise ratio is 0dB, compared with the traditional Costas loop at a carrier frequency offset of 10kHz, the traditional Costas carrier synchronization method takes 27.7 milliseconds to complete carrier synchronization, while the present invention takes 3.4 milliseconds to complete carrier synchronization.

[0012] 2. The present invention adopts a spread spectrum Costas loop carrier fine synchronization method, which greatly improves the receiver's carrier synchronization effect under extremely low signal-to-noise ratio conditions. For example, under a signal-to-noise ratio of -10dB, the traditional Costas loop has a receiving bit error rate of 0.4, while the present invention has a receiving bit error rate of 0.15. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is a system structure diagram of the present invention.

[0014] Figure 2 This is the PMF-FFT structure diagram of the present invention.

[0015] Figure 3 It is a structural diagram of the spread spectrum Costas of the present invention.

[0016] Figure 4 It is a specific flow chart of step 1 of a specific embodiment of the present invention.

[0017] Figure 5 It is a specific flow chart of step 2 of a specific embodiment of the present invention.

[0018] Figure 6 It is a specific flow chart of step 3 of a specific embodiment of the present invention.

[0019] Figure 7 This is a graph showing the bit error rate performance of the present invention compared with the traditional Costas under different signal-to-noise ratios.

[0020] Figure 8 This is a graph showing the synchronization speed of the present invention compared to the traditional Costas at different frequency offsets. DETAILED DESCRIPTION

[0021] To facilitate those skilled in the art to understand the technical content of the present invention, the present invention is further explained below with reference to the accompanying drawings.

[0022] The present application provides a fast synchronization receiver system based on the maximum frequency deviation estimation algorithm of PMF-FFT and improved Costas loop carrier synchronization. By adopting the PMF-FFT algorithm for frequency deviation estimation, the algorithm quickly and accurately determines the frequency deviation of the signal through FFT transformation and partial matching filter estimation. Next, the frequency deviation information is input into the Costas loop for carrier synchronization. The Costas loop achieves high-precision carrier recovery through phase detection, filtering and voltage-controlled oscillator adjustment. In order to improve performance in a low signal-to-noise ratio environment, the system introduces the despread output as a feedback signal, enhances the synchronization capability of the Costas loop, reduces the error of carrier synchronization and further reduces the bit error rate. When the carrier frequency deviation is 10kHz, the bit error rate performance of the present invention compared with the traditional Costas under different signal-to-noise ratios is as shown in the figure. Figure 7 For example, when the signal-to-noise ratio is -10dB, the traditional Costas loop has a receiving bit error rate of 0.4, while the present invention has a receiving bit error rate of 0.15. The system structure diagram is shown in FIG. Figure 1 .

[0023] In the PMF-FFT algorithm module, the received IQ (In Phase & Quadrature Phase) signal first passes through a bandpass filter to remove out-of-band noise. The signal then passes through a squarer to offset the baseband spectrum interference caused by the code elements. The signal then passes through a pulse shaping filter to remove the noise generated by the squarer.

[0024] By extracting the peak value of the sampling points within the FFT calculation period through PMF, the maximum frequency deviation peak value can be calculated more accurately.

[0025] The peak value after FFT (fast Fourier transform) calculation is the maximum frequency deviation range, and the frequency deviation is fed back to the NCO (numerical controlling oscillator, NCO) loop.

[0026] In the despreading loop of the Costas and DSSS systems, the mixing signal generated by the NCO is mixed with the output signal of the low-pass filter in the PMF-FFT preprocessing loop to offset some frequency deviation. The despreading module tracks the mixed signal using the sliding correlation method. Since the spread spectrum sequence has excellent cross-correlation characteristics, the receiver can recover information and suppress noise at very low signal-to-noise ratios. Synchronizing the Costas loop with the despread data can greatly improve tracking speed and stability.

[0027] The phase detector performs phase error identification on the despread signal, and after passing through the loop filter, stable NCO control parameters can be obtained.

[0028] The NCO's control parameters are composed of a PMF-FFT maximum frequency deviation estimation system and an improved Costas loop synchronization tracking system based on spread spectrum. The PMF-FFT can calculate larger frequency deviations to generate fast control parameters, ensuring that the system can quickly complete coarse carrier synchronization. The improved Costas loop synchronization tracking system based on spread spectrum can calculate precise frequency deviations to generate accurate and stable control parameters, ensuring that the system can complete precise carrier synchronization.

[0029] The specific synchronization process is as follows:

[0030] The information is transmitted using spread spectrum modulation, and the code elements are mapped to the [-1, 1] channel, which is an additive white Gaussian noise channel. The received signal is converted into a digital signal after AD conversion. The digital signal is then converted to baseband through digital down-conversion, becoming a baseband digital signal with frequency deviation. The IQ of this baseband signal can be expressed as:

[0031] S[n]=AM[nT s ]G[nT s ]cos(Δω(n)nTs )+N[nT s ]

[0032] Where A is the amplitude of the received signal, T s is the sampling time, Δω(n) is the angular frequency difference between the received signal and the local NCO at the current moment, G[nT s ] is the spreading code at the sampling time, N[nT s ] is the Gaussian noise sampling signal, M[nT s ]Data modulation signal.

[0033] The following steps are as follows:

[0034] Step 1. Signal preprocessing, square S[n]:

[0035] S 2 [n]=A 2 M 2 [nT s ]G 2 [nT s ]cos 2 (ΔωnT s )+N 2 [nT s ]

[0036] M after the pulse shaping filter 2 [nT s ]=1,G 2 [nT s ]=1, we have:

[0037] S 2 [n]=A 2 cos 2 (ΔωnT s )+N 2 [nT s ]

[0038] From this, we can see that after the pulse shaping filter, only the 2-fold frequency deviation component 2Δω and the frequency components in the noise remain. These components are then fed into the PMF. After PMF filtering, an FFT transform is performed, and the transformed signal spectrum is analyzed. The peak value output of the matched filter is squared and then an FFT is performed. The PMF output is:

[0039]

[0040] Among them, n and k represent the serial numbers of the sampled signals, c[n] is the pseudo-random sequence of the spreading code, and the peak output of PMF can be regarded as the output of a despreading method. During the despreading process, the spectrum of the interference signal is broadened by the spreading sequence, so that the interference signal power is evenly distributed over the entire frequency band. The processing process of step 1 is as follows: Figure 4 shown.

[0041] Step 2. PMF and FFT operation, PMF-FFT structure diagram is as follows Figure 2 The FFT algorithm has the characteristic of fast response speed when estimating signal frequency deviation. The number of FFT calculation points has a direct impact on the calculation time and calculation resolution. The more FFT calculation points, the higher the resolution of the calculation result, but the longer the calculation time. Conversely, when the FFT calculation points are fewer, the resolution of the calculation result is lower, but the calculation time is shorter. For fast synchronization systems, there are very high requirements for synchronization time, and the number of calculation points needs to be reduced. For high-precision synchronization systems, there are very high requirements for the accuracy of frequency deviation estimation, and the number of FFT calculation points needs to be increased. Calculation accuracy and calculation time have an inverse relationship. To meet system requirements, a trade-off must be made in the number of FFT calculation points.

[0042] Performing an N-point FFT operation on the PMF output signal f[n] yields:

[0043]

[0044] The frequency calculation formula of FFT is:

[0045] F[n]=n·(f s / N)

[0046] Where F[n] is the frequency corresponding to the nth point of the FFT result, n is the number of points calculated by the FFT point, and f s is the sampling rate, and N is the number of FFT calculation points. The collected signal is downsampled and the FFT operation is performed after downsampling. The frequency calculation formula is:

[0047] F[n]=n·(f s / (MN))

[0048] Where M is the downsampling point interval, and the resolution of the calculation result is increased by M times.

[0049] At a sampling rate of f s =30.72MHz, sampling interval is ΔT=128T s =128 / f s , when the number of FFT calculation points is fftn=512, the calculation frequency range is [-120KHz, 120KHz], the frequency calculation resolution is Δf=1 / (fftnΔT)=468.75Hz, and the total sampling time in a single FFT operation is about 2.1ms. If the signal is processed by FFT, assuming that the position where the maximum value of |F[n]| is located is n=k m , then the maximum frequency offset rough estimate is It can be expressed as:

[0050]

[0051] The frequency offset estimation algorithm based on PMF-FFT can calculate the approximate frequency offset range, but the accuracy of the estimation result is not high. For example, the accuracy of the frequency offset range calculated in the above example is 468.75Hz. This frequency offset range will still affect the performance of the receiver, so it is necessary to accurately estimate and compensate for this offset frequency. The processing process of step 2 is as follows: Figure 5 shown.

[0052] Step 3. Despread the received signal and synchronize the Costas carrier. The loop NCO is controlled by PMF-FFT and Costas phase detection results. The carrier synchronized data is sent to the despreading module, and the despread sliding correlation result is output to the NCO through the BPSK (Binary Phase Shift Keying) phase detector. The principle block diagram is shown as follows: Figure 3 .

[0053] When the signal-to-noise ratio of the received signal is very low, directly using it as a calculation parameter of Costas cannot obtain a good calculation result. The present invention designs a design scheme combining direct sequence spread spectrum communication with Costas.

[0054] The input signal of the direct sequence despreading system is the output after carrier synchronization. The input signal x c [n] is:

[0055] x c [n]=I[n]cosφ+Q[n]sinφ

[0056] Where φ is the signal phase, calculate the input signal x c The correlation between [n] and the locally generated pseudo-random sequence c[n], the sliding correlation function is defined as:

[0057] R[τ]=R[τ-1]+x c [n]·c[n-τ]

[0058] R[τ] is the correlation result. The sliding window moves at different τ values ​​to find the value of n that maximizes R[τ]. Finally, the despread correlation value is output to the Costas loop for carrier synchronization. The despread output r[n] is:

[0059] r[n]=R[n]

[0060] The sliding correlation method effectively detects synchronization of the despreading module's input signal and is well-suited for implementation in logic devices. The sliding correlation acquisition algorithm utilizes the autocorrelation characteristics of pseudorandom codes for correlation detection. When the time difference Δt between the pseudorandom code generated by the receiver and the transmitter is zero, the correlation result is the peak of the autocorrelation function, which represents the sequence length n. When Δt ≠ 0, the correlation result drops sharply to near zero. This characteristic allows for determining whether the receiver is synchronized with the transmitter. If the correlation value falls below a set threshold, indicating a lack of synchronization between the transmitter and receiver, the phase of the local controlled clock is altered, and a T / 2 step is performed. Correlation calculations are then repeated until a correlation peak is found, confirming synchronization. The set threshold here represents the pseudorandom code synchronization determination threshold in spread spectrum communications. This threshold value is conventional and will not be further described in detail in this disclosure.

[0061] Let the real part of r[n] be I[k] and the imaginary part be Q[n], then,

[0062] I[k]=i[k]cos(2πf d kT s +θ n )

[0063] -q[k]sin(2πf d kT s +θ n )

[0064] Q[k]=q[k]cos(2πf d kT s +θ n )

[0065] +i[k]sin(2πf d kT s +θ n )

[0066] Where f d is the Doppler shift, θ n is the random phase caused by the channel, i[k] is the in-phase component of the demodulated output codeword integration, q[k] is the orthogonal component of the demodulated output codeword integration, T s =1 / f s is the sampling period, and through the Costas feedback loop we can get:

[0067] U I [k]=I[k]cosφ+Q[k]sinφ

[0068] =i[k]cos(2πf d kT s +θ n -φ)

[0069] -q[k]sin(2πf d kT s +θ n -φ)

[0070] U q [k]=Q[k]cosφ-I[k]sinφ

[0071] =q[k]cos(2πf d kT s +θ n -φ)

[0072] +i[k]sin(2πf d kT s +θ n -φ)

[0073] In the formula θ e is the phase error. Frequency offset residual and phase error Δφ=2πf d kT s +θ n -φ, we know that when Δφ≈0, cos(Δφ)=1, sin(Δφ)=0, we can get the phase detector error U pd [k]:

[0074] U pd [k]=U q [k]·sgn{U i [k]}

[0075] ={q[k]cos(Δφ)+i[k]sin(Δφ)}·sgn(i[k])

[0076] When the loop is in tracking state, the frequency offset residual and phase error Δφ approach 0, then:

[0077] U pd [k] = q[k]·sgn(i[k]) = 0

[0078] Loop to U pd [k] is used for detection and discrimination. At the beginning of synchronization, the residual frequency difference is large and the loop is tracking quickly. pd When [k] gradually approaches 0, it is considered that the synchronization is successful and the lock is performed. The processing process of step 3 is as follows Figure 6 shown.

[0079] In the present invention, the PMF-FFT quickly tracks and locks the coarse frequency deviation of the received signal. The NCO controls the signal frequency deviation to quickly drop to a small range based on the feedback results of the PMF-FFT. Spread spectrum synchronization can recover baseband information and filter out noise under extremely low signal-to-noise ratio conditions. The Costas loop uses the spread spectrum output results as detection parameters to quickly complete fine carrier synchronization.

[0080] When the signal-to-noise ratio of the received information is 0dB, the synchronization speed performance of the present invention compared with the traditional Costas loop at a carrier frequency deviation of 10kHz is shown in the figure below. Figure 8 The traditional Costas carrier synchronization method takes 27.7 milliseconds to complete carrier synchronization, while the present invention takes 3.4 milliseconds to complete carrier synchronization.

[0081] Those skilled in the art will appreciate that the embodiments described herein are intended to aid the reader in understanding the principles of the present invention, and it should be understood that the scope of the present invention is not limited to such specific descriptions and embodiments. Various modifications and variations are readily apparent to those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention are intended to be included within the scope of the claims.

Claims

1. A DSSS enhanced synchronization method combining PMF-FFT and improved Costas loop, characterized in that: The frequency offset estimation process is implemented using a PMF-FFT algorithm module, which includes: a bandpass filter, a squarer, a pulse shaping filter, a PMF operation unit, and an FFT operation unit connected in sequence; the received IQ signal is passed through the PMF-FFT algorithm module to obtain a rough estimate of the maximum frequency offset; and a low-pass filter, the input of which is the output of the bandpass filter; The carrier synchronization process is implemented using an improved Costas loop based on a PMF-FFT loop. The improved Costas loop includes: a despreading module, a phase detector, a loop filter, and a digitally controlled oscillator. The inputs of the despreading module are the output of the low-pass filter in the PMF-FFT algorithm module and the output of the digitally controlled oscillator. The output of the despreading module serves as the input of the phase detector, and the output of the phase detector serves as the input of the loop filter. The output of the loop filter and the coarse maximum frequency offset estimate serve as the input of the digitally controlled oscillator: When the phase detector error is 0, locking is performed and a synchronized demodulated output is obtained.

2. The DSSS enhanced synchronization method combining PMF-FFT and improved Costas loop according to claim 1, characterized in that: The method also includes down-sampling the output signal of the PMF operation unit.

3. The DSSS enhanced synchronization method combining PMF-FFT and improved Costas loop according to claim 2, characterized in that: After the downsampled signal is processed by the FFT operation unit, the frequency calculation formula is: F[n]=n·(f s / (MN)) Where M is the sampling point interval, N is the number of FFT calculation points, and f s is the sampling rate.

4. The DSSS enhanced synchronization method combining PMF-FFT and improved Costas loop according to claim 3, characterized in that: After being processed by the FFT operation unit, the position k where the maximum value of |F[n]| is located is m = k, and the maximum frequency offset rough estimate is obtained. The maximum frequency offset rough estimate is recorded as but The corresponding calculation formula is:

5. The DSSS enhanced synchronization method combining PMF-FFT and improved Costas loop according to claim 4, characterized in that: The despreading module uses a sliding correlation method to detect whether the input signal of the despreading module is synchronized.

6. The DSSS enhanced synchronization method combining PMF-FFT and improved Costas loop according to claim 5, characterized in that: The demodulated output after synchronization is expressed as: r[n]=R[n] Where r[n] represents the despread output, and R[n] represents the maximum value of the sliding correlation function.

Images