A terahertz communication and sensing integrated system under large Doppler frequency deviation

Abstract

The present invention belongs to the field of terahertz communication technology, and specifically relates to an integrated terahertz communication perception system under a large Doppler frequency deviation. The present invention uses an additional OFDM symbol as a pilot and places it in the frame header, and this additional OFDM symbol carries only one subcarrier (N△f / 2) data. The receiving end obtains a rough estimate of the Doppler by analyzing the non-origin DFT of this symbol, and obtains a fine estimate of the Doppler by using the traditional OFDM perception algorithm after compensation, and finally synthesizes the estimated Doppler and speed. This scheme ensures that the speed range is expanded by about N times while the speed resolution remains unchanged, solves the ambiguity problem of speed estimation when there is a large Doppler frequency deviation, and can also obtain a rough estimate of the frequency deviation at the communication end.

Description

Technical Field

[0001] The present invention belongs to the technical field of terahertz communication, and in particular relates to a terahertz communication sensing integrated system under a large Doppler frequency deviation. Background Art

[0002] The terahertz band is between microwave and infrared visible light bands, with a frequency range of 0.1THz to 10THz (corresponding to a wavelength of 30 to 3000μm), and is an intersection of electronics and photonics. This band has broad spectrum resources and can provide high-speed, reliable, and low-latency communication services for more user devices. Compared with the millimeter wave band, the terahertz band has a larger bandwidth and a smaller wavelength, and is particularly suitable for use in medium and short-range high-precision communication perception scenarios. At the same time, the small wavelength characteristics make it easier to integrate enough antennas in miniaturized devices, thereby promoting the popularization and application of synaesthesia integrated devices. From a perception perspective, the terahertz band has a wide bandwidth and enough antennas, so high-precision positioning and imaging applications can be achieved in close-range scenarios. Therefore, the future development prospects of terahertz-based ISAC technology are very broad. In traditional OFDM-ISAC systems, due to the low carrier frequency and small target speed, the Doppler frequency deviation generated is usually within a subcarrier spacing △f. However, when applied to terahertz and high-speed mobile scenarios, the resulting Doppler frequency deviation may exceed a subcarrier spacing △f, which will not only affect communication but also cause ambiguity in speed detection. Summary of the invention

[0003] The present invention provides a terahertz communication and perception integrated system under a large Doppler frequency deviation.

[0004] The technical solution adopted by the present invention is that, on the basis of OFDM, the transmitting end considers adding a pilot (time domain OFDM symbol), and at the receiving end, firstly performs DFT analysis on the pilot with the non-origin as the starting point to obtain a rough estimate of the frequency deviation. Then, the rough estimate of the frequency deviation is compensated to the received signal, and the frequency deviation (speed) estimation method of the traditional OFDM is used for it. Finally, the two are combined to form a joint estimation method for speed.

[0005] The technical solution of the present invention is:

[0006] A terahertz communication and sensing integrated system under Doppler frequency deviation is proposed. The system is defined as a single-antenna system for both the transmitter and the receiver, with only a single target, a narrowband terahertz communication and sensing channel, N subcarriers, and a subcarrier spacing of △f. Considering the Doppler frequency deviation |f D | <N△f / 2;

[0007] Transmitter:

[0008] Based on the OFDM transmission signal, a pilot symbol is added to the frame header:

[0009]

[0010] in, It is used to ensure that the power of the pilot symbol is the same as that of the subsequent transmitted symbol. QAM represents the total number of constellation points when QAM symbols are mapped, k represents the sampled index number of an OFDM basic symbol, and T is the length of the OFDM basic symbol;

[0011] Receiver:

[0012] Assume the Doppler frequency deviation is:

[0013] f D =f D,rough +f D,tiny =e△f+f tenths +f D,tiny ,e∈Z

[0014] Where Z represents an integer; f D,rough represents the frequency deviation value of the Doppler rough estimate obtained by the DFT algorithm using a non-origin as the starting point, f D,tiny It represents the small frequency deviation value remaining after the rough estimation compensation. tenths ∈{-0.5△f,-0.4△f,…,0,…,0.4△f}, f D,tiny ∈[0,△f), let the speed corresponding to Doppler be:

[0015]

[0016] Based on the OFDM received signal, the Doppler frequency offset is roughly estimated through the pilot symbol. Specifically, the pilot symbol p(k) is subjected to the Doppler frequency offset f D The resulting signal q(k) is subjected to DFT analysis using a non-origin point as the starting point to obtain a rough estimate of the Doppler frequency offset f D,rough , and its estimation accuracy is 0.1△f; the DFT analysis with non-origin as the starting point refers to the DFT analysis with the starting point as f 0 =-0.5△f,-0.4△f,…,0,…,0.4△f Perform DFT analysis on q(k) with non-origin as the starting point:

[0017]

[0018] Compare the peak values ​​in each DFT result, and take the horizontal coordinate corresponding to the largest peak value as the corresponding value of Doppler frequency deviation:

[0019]

[0020] Considering that the Doppler frequency deviation exceeds the range of a subcarrier spacing △f, the rough estimation value of the Doppler frequency deviation f is finally obtained. D,rough for:

[0021]

[0022] The rough estimate of Doppler frequency offset f D,rough Compensate the received baseband signal and perform OFDM demodulation, which is equivalent to obtaining a distance of R and a speed of v tiny The echo baseband signal:

[0023]

[0024] Where b represents the attenuation coefficient, M represents the number of OFDM symbols, S(μ,n) represents the data carried on the n+1th subcarrier of the μ+1th OFDM symbol, and T cp Indicates the time length of CP; T s represents the total OFDM symbol length, z(t) represents additive white Gaussian noise;

[0025] For s (t) Sampling and performing FFT operation to convert it into frequency domain:

[0026]

[0027] in, Z(μ,n) represents the noise term;

[0028] To estimate the Doppler frequency deviation: s (μ,n) is divided element by element by the frequency domain symbol S(μ,n) during OFDM modulation at the transmitter to obtain the perceived channel matrix H s :

[0029]

[0030] Among them, H s (μ,n) represents H s The μ+1th column and n+1th row of , Z(μ,n) / S(μ,n) represents the noise floor that depends on digital modulation; for H s (μ,n) performs M-point FFT along the OFDM symbol direction and recovers the velocity information independently:

[0031]

[0032] in, represents the noise part;

[0033] when When |V(p,n)| reaches its peak value, the horizontal coordinate p′ when |V(p,n)| reaches its peak value is converted into the velocity scale to obtain a detailed estimate of the velocity:

[0034]

[0035] The final target velocity estimate is given by:

[0036]

[0037] The beneficial effects of the present invention are as follows: the present invention uses an additional OFDM symbol as a pilot signal placed in the frame header, and the additional OFDM symbol carries only one subcarrier (N△f / 2) data. The receiving end obtains a rough estimate of the Doppler by analyzing the non-origin DFT of the symbol, and obtains a fine estimate of the Doppler by using the traditional OFDM sensing algorithm after compensation, and finally synthesizes the estimated Doppler and speed. This scheme ensures that the speed range is expanded by about N times while the speed resolution remains unchanged, solves the ambiguity problem of speed estimation when there is a large Doppler frequency deviation, and can also obtain a rough estimate of the frequency deviation at the communication end. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 The pilot frequency domain provided by the present invention has only one subcarrier carrying data;

[0039] Figure 2 It is a schematic diagram of non-origin DFT analysis provided by the present invention;

[0040] Figure 3 1 is a comparison diagram between the conventional OFDM-based perception provided by the present invention and the proposed method. (a)-(c) are conventional perception speed / distance diagrams at speeds of 40 m / s, 110 m / s and 2000 m / s and a distance of 800 m; (d)-(f) adopt the algorithm proposed by the present invention, and the calculated frequencies of the algorithm are 60 kHz, 156 kHz and 2.94 MHz respectively;

[0041] Figure 4 It is an MSE diagram of the non-origin DFT+CP-based normalized frequency offset estimation provided by the present invention. DETAILED DESCRIPTION

[0042] The present invention is further described in detail below in conjunction with the accompanying drawings and embodiments.

[0043] The present invention proposes a DFT method with a non-origin as the starting point, and the Doppler frequency offset estimation accuracy can reach 0.1△f. The DFT algorithm with a non-origin as the starting point can be used not only for the rough estimation of the frequency offset at the communication end, but also for the rough estimation of the Doppler frequency offset at the sensing end, and converted into a rough estimation of the target speed.

[0044] Conventional DFT analysis is to perform N-point DFT calculation on the received signal, which is equivalent to sampling at △f in the frequency domain and calculating the Fourier level values ​​at these sampling points. Non-origin DFT analysis is when the signal is sampled in the frequency domain, the sampling points are not sampled from the origin, but at a certain offset f. 0 As the starting point of the sampling point, the Fourier level values ​​calculated after multiple different offset samplings constitute a DFT analysis with a frequency interval of 0.1△f. Therefore, the DFT analysis using a non-origin as the starting point can roughly locate the Doppler spectrum but more accurately than the conventional DFT algorithm, such as Figure 2 shown.

[0045] Take the starting point as f 0 =-0.5△f,-0.4△f,…,0,…,0.4△f Performing DFT analysis on q(k) in equation (1-2) with non-origin as the starting point, we have

[0046]

[0047] And compare the peak values ​​in each DFT result, and take the horizontal coordinate corresponding to the largest peak value as the corresponding value of Doppler frequency deviation, that is,

[0048]

[0049] We also need to consider that the Doppler frequency deviation exceeds the range of a subcarrier spacing △f, then we have

[0050]

[0051] The fft function in the MATLAB toolkit uses DFT analysis with 0 as the starting point, so you also need to redesign the DFT toolkit for non-origin points.

[0052] According to the frequency shift characteristics of Fourier transform: the shift in the frequency domain is equivalent to the phase shift in the time domain, and the phase shift factor is Therefore, the final implementation is as follows:

[0053]

[0054] In this embodiment, f c =220GHz, N=8192, M=512, △f=120kHz, single target distance R=800m and speed v=40, 110, 2000m / s respectively were used as examples for simulation. Figure 3 shown.

[0055] In the communication receiver, a combination of DFT coarse frequency offset estimation with a non-origin as the starting point and CP-based fine frequency offset estimation is adopted, and finally the mean square error (MSE) of normalized frequency offset estimation (normalized frequency offset is defined as the ratio of frequency offset to subcarrier spacing) can be achieved to be less than 10-5, and frequency offsets greater than the subcarrier spacing can also be estimated. The MSE calculated by simulation is as follows: Figure 4 As shown, the relative motion speed of the simulation setting is 523.2741m / s.

Claims

1. A terahertz communication and sensing integrated system under large Doppler frequency offset, where both the transmitter and receiver in the system are single antennas and there is only a single target. The channel is a narrowband terahertz communication and sensing channel, the number of subcarriers is N, and the subcarrier spacing is Δf. Considering the Doppler frequency offset |f D | < NΔf / 2, and it is characterized in that: Transmitter: Based on the OFDM transmission signal, a pilot symbol is added to the frame header: in, It is used to ensure that the power of the pilot symbol is the same as that of the subsequent transmitted symbol. QAM represents the total number of constellation points when QAM symbols are mapped, k represents the sampled index number of an OFDM basic symbol, and T is the length of the OFDM basic symbol; Receiver: Based on the OFDM received signal, the Doppler frequency offset is roughly estimated through the pilot symbol. Specifically, the pilot symbol p(k) is subjected to the Doppler frequency offset f D The resulting signal q(k) is subjected to DFT analysis using a non-origin point as the starting point to obtain a rough estimate of the Doppler frequency offset f D,rough ; The DFT analysis with a non-origin as the starting point refers to the DFT analysis of q(k) with a starting point of f0=-0.5△f,-0.4△f,…,0,…,0.4△f as the starting point: Compare the peak values ​​in each DFT result, and take the horizontal coordinate corresponding to the largest peak value as the corresponding value of Doppler frequency deviation: Considering that the Doppler frequency deviation exceeds the range of a subcarrier spacing △f, the rough estimation value of the Doppler frequency deviation f is finally obtained. D,rough for: The rough estimate of Doppler frequency offset f D,rough Compensate the received baseband signal and perform OFDM demodulation, which is equivalent to obtaining a distance of R and a speed of v tiny The echo baseband signal: Where b represents the attenuation coefficient, M represents the number of OFDM symbols, S(μ,n) represents the data carried on the n+1th subcarrier of the μ+1th OFDM symbol, and T cp Indicates the time length of CP; T s represents the total OFDM symbol length, z(t) represents additive white Gaussian noise; For s (t) Sampling and performing FFT operation to convert it into frequency domain: in, Z(μ,n) represents the noise term; To estimate the Doppler frequency deviation: s (μ,n) is divided element by element by the frequency domain symbol S(μ,n) during OFDM modulation at the transmitter to obtain the perceived channel matrix H s : Among them, H s (μ,n) represents H s The μ+1th column and n+1th row of , Z(μ,n) / S(μ,n) represents the noise floor that depends on digital modulation; for H s (μ,n) performs M-point FFT along the OFDM symbol direction and recovers the velocity information independently: in, represents the noise part; when When V(p,n) reaches its peak value, the horizontal coordinate p′ when V(p,n) reaches its peak value is converted into the velocity scale to obtain a detailed estimate of the velocity: The final target velocity estimate is given by:

Images