CN104378319A - Channel estimation method based on short wave channel MIMO-OFDM communication system - Google Patents

Abstract

The invention discloses a channel estimation method based on a short-wave channel MIMO-OFDM communication system. A pilot sequence composed of cyclic orthogonal codes is constructed at the sending end, and is intermittently inserted into sending data to form a time slot of a double-cycle structure. structure, where the pilot sequence of the mth transmitting antenna is the pilot sequence of the 0th transmitting antenna and is cyclically shifted to the right by m×D _max bits, where D _max is the maximum time delay of the channel, and the time slot is divided into multiple sub-times The guard segment of each sub-slot is composed of the last L _G symbols of the pilot, and the last L _G + L _p symbols of the entire time slot form the guard segment and the pilot segment at the beginning of the slot; For each pilot segment of the signal, the channel impulse response is estimated by the least square method, and then the time correlation is used to obtain the channel impulse response parameters within the data segment. The method of the invention improves the accuracy of channel parameter estimation and reduces the complexity of realization, thereby promoting the improvement of system performance.

Description

A Channel Estimation Method Based on HF Channel MIMO-OFDM Communication System

technical field

The invention relates to a channel estimation method based on a shortwave channel MIMO-OFDM system, which belongs to the field of communication system receiver channel estimation.

Background technique

Short-wave communication (also known as high-frequency communication) is to achieve medium and long-distance communication through ionospheric reflection or short-distance communication transmission through ground waves. The actual frequency range used is 3MHz-30MHz. It has the advantages of simple equipment, low cost, high flexibility, and long communication distance, and is an indispensable type of wireless communication. Short-wave transmission mainly relies on ionospheric reflection. Since the ionosphere is a layered, inhomogeneous, anisotropic, and time-space medium, short-wave channels have multipath delays, fading, Doppler frequency deviation, and frequency dispersion. A series of complex features.

MIMO (Multiple Input Multiple Output) systems use multiple antennas to transmit and receive signals, which can increase channel capacity and have a higher frequency spectrum than SISO (Single Input Single Output) systems using the same total power and bandwidth. utilization rate. Under ideal conditions, the capacity of a MIMO system increases linearly with the number of antennas. At the same time, the emergence of Orthogonal Frequency Division Multiplexing (OFDM) technology makes the data transmission rate on each subcarrier relatively low, thereby reducing the intersymbol interference. This technology makes the system insensitive to the delay dispersion of the channel, and can obtain good performance even without using an equalizer. Applying MIMO and OFDM technology to the short-wave communication system can effectively improve the performance of the short-wave communication system. Therefore, constructing MIMO-OFDM system based on shortwave channel is a problem worth studying at present.

In MIMO-OFDM systems based on shortwave channels, due to a series of influences such as multipath delay, fading, Doppler frequency shift/Gaussian white noise in wireless channels, it is a difficult process to estimate channel parameters at the receiving end. Commonly used non-blind channel estimation uses reference signals, such as pilots or training sequences, to estimate channel parameter values according to certain estimation criteria, and at the same time gradually track and adjust the estimated values of the parameters to be estimated. In such an algorithm, the design of the OFDM time slot structure and the improvement of the pilot (sequence) scheme will affect the result of channel estimation and the implementation complexity, resulting in changes in system performance.

Contents of the invention

Purpose of the invention: for MIMO-OFDM systems based on shortwave channels, the present invention provides a channel estimation method, which adopts a dual-cycle adaptive time slot structure and a pilot scheme composed of a set of cyclic orthogonal code cyclic shifts, to The accuracy of channel parameter estimation is improved and the complexity of implementation is reduced.

Technical solution: In order to achieve the above-mentioned purpose of the invention, the present invention adopts the following technical solution:

A channel estimation method based on a shortwave channel MIMO-OFDM communication system, comprising the steps of:

(1) At the sending end, construct a pilot sequence composed of cyclic orthogonal codes, and insert it intermittently into the sending data to form a double-cycle time slot structure, wherein the pilot sequence of the mth transmitting antenna is The pilot sequence of 0 transmitting antennas is cyclically shifted to the right by m×D _max bits, 0<m<M, M is the number of transmitting antennas, and D _max is the maximum time delay of the channel; the time slot structure consists of multiple A time slot and a tail sequence, each sub-slot is composed of a cyclic protection segment G, a pilot segment P, and a user data segment D or control information C, and the tail sequence is composed of G+P; the cyclic protection segment G The length L _G is not less than the maximum delay of the channel, and is less than the length L _p of the pilot segment P; the cyclic protection segment G is composed of the last L _G symbols of the adjacent pilot segment P;

(2) At the receiving end, use each pilot segment of the received signal to estimate a set of channel impulse response estimates using the least squares method in is the transpose of the matrix S _p composed of pilot signals on all transmitting antennas, r _P,n is the pilot sequence received by the nth receiving antenna;

(3) At the receiving end, on the basis of the channel impulse response estimated in step (2), the time correlation is used to obtain the channel impulse response parameters within the data period.

Beneficial effects: the time slot structure adopted in the channel estimation method of the present invention can eliminate the interference between each time slot, and also eliminate the interference of the unknown data symbols in the time slot to the determined pilot sequence, and only need to take out the pilot sequence when performing channel estimation. Band processing is fine. This special time slot form can transform the linear convolution of the channel impulse response and the transmitted signal sequence into a circular convolution, so that the received vector can be simply expressed by the product of the channel circular matrix and the transmitted signal vector. The adopted pilot scheme can guarantee the orthogonality of the pilot symbols on each antenna, and can simplify the calculation of the least squares method in the MIMO-OFDM system based on the shortwave watterson model, making the inversion operation in the algorithm simple Matrix transpose and division forms.

Description of drawings

Fig. 1 is a MIMO-OFDM communication system model diagram;

Fig. 2 is the structural diagram of the system time slot that the present invention adopts;

Fig. 3 is a diagram of a pilot sequence formed by cyclic shifting of the cyclic orthogonal code adopted in the present invention.

Detailed ways

The present invention will be further elaborated below in conjunction with specific embodiments.

Fig. 1 is the system model diagram based on the MIMO-OFDM communication system of shortwave channel, the channel estimation method that is applied to this system of the present invention, comprises the steps:

(1) The transmitting end constructs a pilot sequence according to the format in FIG. 3 , and inserts the pilot sequence intermittently into the transmission data to form a time slot structure with a double cycle structure as shown in FIG. 2 . The time slot structure is composed of 4 sub-slots and a tail sequence. Each sub-slot is composed of a cyclic protection segment G, a pilot segment P, a user data segment D or a control information C. The tail sequence of a time slot is G+P. The length L _G of the cyclic guard segment should not be less than the maximum delay of the channel, and at the same time be less than the pilot length L _p . The time slot structure adopts a double-cycle adaptive structure, that is, for G+P, G is composed of the last L _G symbols of P, which is a small cycle; and for the entire time slot, (G+P) and The last L _G + L _p symbols of the entire time slot are consistent, which is a big cycle. The advantage of this time slot structure is that it can eliminate the interference between time slots, and at the same time eliminate the interference of unknown data symbols in the time slot to the determined pilot sequence. When performing channel estimation, it is only necessary to take out the pilot frequency segment for processing; The linear convolution of the channel impulse response and the transmitted signal sequence is transformed into a circular convolution, so that the received vector can be simply expressed by the product of the channel circular matrix and the transmitted signal vector. The pilot sequence shown in Figure 3, wherein j is an imaginary symbol, the length of the pilot P is 32 symbols, and each symbol is a complex number. The length of the cyclic protection segment G is 8, and is composed of the last 8 symbols of each pilot segment P. In order to take advantage of the cyclic orthogonality of the pilot sequence, the pilot sequence of the mth transmitting antenna is cyclically shifted to the right by mP bits from the pilot sequence of the 0th transmitting antenna (P is the maximum time delay of the channel).

(2) At the receiving end, use each pilot segment of the received signal to estimate a group of channel impulse responses by using the least square method.

Under the multipath time-varying channel, the complex baseband signal of the nth receiving antenna at the receiving end can be expressed as:

${\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}\underset{\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; the\; s</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>$

where s _m (k) is the digital baseband signal sent by the mth antenna at the transmitter, h _n,m (l,k) is the time-varying impulse response from the mth antenna to the nth antenna, D _max is the channel The maximum time delay, z _n (k) is a zero-mean white Gaussian noise sequence. In the shortwave channel modeled by Watterson, it can be considered that the channel is nearly time-invariant in the time period occupied by each sub-slot, and therefore, in the discussion involving only the data sequence processing of the length of a single sub-slot, it will be omitted The second argument in h _n,m (l,k). According to the received pilot signal in each time slot, the impulse response of the channel and the variance of the noise can be estimated for subsequent signal detection.

Since cyclic protection is added before each pilot segment of length L _P in the transmitted signal, and the length of the cyclic protection sequence is not less than the maximum delay D _max of the channel, each receiving antenna receives a cyclic protection sequence of length L _P The pilot sequence can be expressed as the superposition of the circular convolution of the transmission pilot sequence of M transmitting antennas and the channel impulse response sequence, plus the zero-mean Gaussian white noise sequence. Let s _P,m (k), (k=0,1,...,L _P -1) represent the transmission pilot sequence of the mth antenna, and r _P,n (k), (k=0, 1,..., _LP -1) represents the pilot sequence received by the nth receiving antenna, then:

$\begin{array}{cc}{\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; the\; s</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&CircleTimes;</span>{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span>& <span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0,1</span>}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; L</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>\\ & <span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0,1</span>}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>\end{array}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span>$

in Represents circular convolution, h _n,m (l), (l=0,1,...,D _max -1) represents the channel impulse response sequence from the mth transmitting antenna to the nth receiving antenna, z _{P , n} (k) is a zero-mean Gaussian white noise sequence, and its variance is set as To estimate the channel impulse response from each transmit antenna to each receive antenna, we can make:

h _n,m =[h _n,m (0)h _n,m (1)…h _n,m (D _max -1)] ^T (3)

${\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; =</span>{\left[\begin{array}{cccc}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 0</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}& {\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}& <span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span>& {\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}\end{array}\right]}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 4</span>}<span\; class="notranslate">\; )</span>$

s _P,m =[s _P,m (0) s _P,m (1) … s _P,m (L _P -1)] ^T (5)

z _P,n =[z _P,n (0) z _P,n (1) … z _P,n (L _P -1)] ^T (6)

In order to take advantage of the cyclic orthogonality of the pilot sequence, the pilot sequence of the mth transmit antenna is cyclically shifted to the right by m*D _max bits for the pilot sequence of the 0th transmit antenna, namely:

${\mathrm{<span\; class="notranslate">\; the\; s</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; =</span>\left[\begin{array}{cc}\mathrm{<span\; class="notranslate">\; 0</span>}& {\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}}\\ {\mathrm{<span\; class="notranslate">\; I</span>}}_{{\mathrm{<span\; class="notranslate">\; L</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; m</span>}{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}}& \mathrm{<span\; class="notranslate">\; 0</span>}\end{array}\right]{\mathrm{<span\; class="notranslate">\; the\; s</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 7</span>}<span\; class="notranslate">\; )</span>$

make in:

${\mathrm{<span\; class="notranslate">\; the\; s</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; m</span>}}^{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; )</span>}<span\; class="notranslate">\; =</span>\left[\begin{array}{cc}\mathrm{<span\; class="notranslate">\; 0</span>}& {\mathrm{<span\; class="notranslate">\; I</span>}}_{{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}}\\ {\mathrm{<span\; class="notranslate">\; I</span>}}_{{\mathrm{<span\; class="notranslate">\; L</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}}& \mathrm{<span\; class="notranslate">\; 0</span>}\end{array}\right]{\mathrm{<span\; class="notranslate">\; the\; s</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 8</span>}<span\; class="notranslate">\; )</span>$

Let S _P ＝[S _P,0 S _P,1 … S _P,M-1 ], then (2) can be expressed as

r _P,n ＝S _P h _n +z _P,n (9)

Use the least squares channel estimation method to obtain the estimated value of the channel parameters

${\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; h</span>}}}_{\mathrm{<span\; class="notranslate">\; LS</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; =</span>{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; P</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; P</span>}}<span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; P</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; P</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 10</span>}<span\; class="notranslate">\; )</span>$

Due to the cyclic orthogonal nature of the pilot sequence, namely:

${\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; P</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; P</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; L</span>}}_{\mathrm{<span\; class="notranslate">\; P</span>}}{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; MP</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 11</span>}<span\; class="notranslate">\; )</span>$

Where I _MP represents the identity matrix of (M*D _max )X(M*D _max ). At this time, the least squares channel estimate of h _n for:

${\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; h</span>}}}_{\mathrm{<span\; class="notranslate">\; LS</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{{\mathrm{<span\; class="notranslate">\; L</span>}}_{\mathrm{<span\; class="notranslate">\; P</span>}}}{\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; P</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; P</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 12</span>}<span\; class="notranslate">\; )</span>$

Under the least squares channel estimation, accurate channel estimation can be obtained by using the pilot sequence with cyclic orthogonal characteristics.

Based on the above derivation results, the calculation steps of the channel parameter estimation algorithm at the receiving end are as follows:

(2.1) Construct the matrix S _P according to the structure of S _P =[S _P,0 S _P,1 ... S _P,M-1 ], where each column vector can be expressed as s _P,m =[s _P,m ( 0) s _P,m (1) … s _P,m (L _P -1)] ^T , indicating the transmit pilot sequence of the mth transmit antenna.

(2.2) Construct r _p,n =[r _p,n (0) r _p,n (1) ... r _p,n (L _p -1)] ^T represents the pilot sequence received by the nth receiving antenna, This sequence is the known received signal.

(2.3) Press ${\hat{h}}_{\mathrm{LS},\mathrm{no}}=\frac{1}{L_P}{S}_P^h{r}_{P,\mathrm{no}}$ Calculate $h_{\mathrm{no}}={\left[\begin{array}{cccc}{h}_{\mathrm{no},0}^T& h_{\mathrm{no},1}^T& ...& h_{\mathrm{no},m-1}^T\end{array}\right]}^T,$ Each column vector is h _n,m =[h _n,m (0) h _n,m (1) ... h _n,m (D _max -1)] ^T . The calculated value at this time is the channel parameter at the pilot based on the least square method.

(3) On the basis of the channel impulse response estimated in step (2), the time correlation is used to obtain the impulse response parameters of the channel within the data period, so as to complete the work of the channel estimation module and provide the follow-up information for the receiver. Detection provides parameters. Since the Watterson model is used as the simulation implementation of the channel, the channel parameters with a short interval between the two pilot segments can be estimated by interpolation methods, such as linear interpolation, Gaussian interpolation, spline interpolation, etc., so as to obtain the corresponding time of the data segment channel parameters.

According to the foregoing derivation, it can be deduced that the time slot structure and pilot parameters used in the embodiment of the present invention can simplify the least squares estimation algorithm in the case of multiple input and multiple output, and ensure the orthogonality of each input and output channel, thereby Improve system performance.

Claims (5)

1. a channel estimation method based on shortwave channel MIMO-OFDM communication system, is characterized in that, comprises the steps: (1) At the sending end, construct a pilot sequence composed of cyclic orthogonal codes, and insert it intermittently into the sending data to form a double-cycle time slot structure, wherein the pilot sequence of the mth transmitting antenna is The pilot sequence of 0 transmitting antennas is cyclically shifted to the right by m×D _max bits, 0<m<M, M is the number of transmitting antennas, and D _max is the maximum time delay of the channel; the time slot structure consists of multiple The time slot is composed of a tail sequence, and each sub-slot is composed of a cyclic protection segment G, a pilot segment P, and a user data segment D or control information C, and the tail sequence is composed of G+P; the length of the cyclic protection segment G is L _G is not less than the maximum channel delay and less than the length L _p of the pilot segment P; the cyclic protection segment G is composed of the last L _G symbols of the adjacent pilot segment P; (2) At the receiving end, use each pilot segment of the received signal to estimate a set of channel impulse response estimates using the least squares method in is the transpose of the matrix S _p composed of pilot signals on all transmitting antennas, r _P,n is the pilot sequence received by the nth receiving antenna; (3) At the receiving end, on the basis of the channel impulse response estimated in step (2), the time correlation is used to obtain the channel impulse response parameters within the data period. 2. the channel estimation method based on shortwave channel MIMO-OFDM communication system according to claim 1, is characterized in that, described utilizes each pilot frequency segment of received signal, adopts least square method to estimate one group of channel impulse response Calculation steps include: (2.1) Construct the matrix S _P according to the structure of S _P =[S _P,0 S _P,1 ... S _P,M-1 ], wherein each column vector is s _P,m =[s _P,m (0) s _P,m (1) … s _P,m (L _P -1)] ^T , indicating the transmit pilot sequence of the mth transmit antenna; (2.2) Construct r _p,n =[r _p,n (0) r _p,n (1) ... r _p,n (L _p -1)] ^T represents the pilot sequence received by the nth receiving antenna, The sequence is a known received signal; (2.3) According to the formula ${\hat{h}}_{\mathrm{LS},\mathrm{no}}=\frac{1}{L_P}{S}_P^h{r}_{P,\mathrm{no}}$ Calculate $h_{\mathrm{no}}={\left[\begin{array}{cccc}{h}_{\mathrm{no},0}^T& h_{\mathrm{no},1}^T& ...& h_{\mathrm{no},m-1}^T]\end{array}\right]}^T$ The least squares channel impulse response estimation of , where each column vector is h _n,m =[h _n,m (0)h _n,m (1)…h _n,m (D _max -1)] ^T . 3. the channel estimation method based on shortwave channel MIMO-OFDM communication system according to claim 1, is characterized in that, adopts linear interpolation, Gaussian interpolation or spline interpolation algorithm to calculate the channel at data point place in the described step (3) Impulse response parameters. 4. the channel estimation method based on shortwave channel MIMO-OFDM communication system according to claim 1, is characterized in that, described cyclic orthogonal code is: $\left[\begin{array}{cccccccc}\mathrm{<span\; class="notranslate">\; 1111</span>}\frac{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; j</span>}}{\sqrt{\mathrm{<span\; class="notranslate">\; 2</span>}}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}& \frac{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; j</span>}}{\sqrt{\mathrm{<span\; class="notranslate">\; 2</span>}}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; j\; j</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; j</span>}& \mathrm{<span\; class="notranslate">\; 1</span>}\frac{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; j</span>}}{\sqrt{\mathrm{<span\; class="notranslate">\; 2</span>}}}& \mathrm{<span\; class="notranslate">\; 1</span>}\frac{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; j</span>}}{\sqrt{\mathrm{<span\; class="notranslate">\; 2</span>}}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}& \mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}\mathrm{<span\; class="notranslate">\; 1</span>}\frac{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; j</span>}}{\sqrt{\mathrm{<span\; class="notranslate">\; 2</span>}}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}& \frac{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; j</span>}}{\sqrt{\mathrm{<span\; class="notranslate">\; 2</span>}}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}\mathrm{<span\; class="notranslate">\; j</span>}& \mathrm{<span\; class="notranslate">\; 1</span>}\frac{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; j</span>}}{\sqrt{\mathrm{<span\; class="notranslate">\; 2</span>}}}& \mathrm{<span\; class="notranslate">\; 1</span>}\frac{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; j</span>}}{\sqrt{\mathrm{<span\; class="notranslate">\; 2</span>}}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; j</span>}\end{array}\right]$ Where j is an imaginary number symbol. 5. the channel estimation method based on short-wave channel MIMO-OFDM communication system according to claim 1, described time slot structure is made of 4 sub-slots and a tail sequence, and the cyclic guard section length of each sub-slot is 8 symbols, the length of the pilot segment is 32 symbols.