CN101697622A - Methods for eliminating intra-cell interference and inter-cell interference in TD-SCDMA system - Google Patents

Abstract

A method for effectively eliminating intra-cell and inter-cell interference in a TD-SCDMA system. TD-SCDMA adopts smart antenna and multi-user joint detection technology, and is a mobile communication standard with independent intellectual property rights. Smart Antenna (Smart Antenna, SA) technology is to dynamically track the desired user's shaped beam to receive and send signals, which greatly reduces the interference between system cells. The data rate of 3G is high and the working characteristics are complex. Smart antennas are required to provide accurate and flexible interference control. Intra-cell interference includes multiple access interference and intersymbol interference. Joint detection (JD) technology is a good anti-multiple access interference and Intersymbol interference technology. The combination of the two can reduce interference, thereby increasing the capacity of the system, obtaining high frequency spectrum utilization, making the power control part in the base station and user equipment simpler, and reducing network costs at the same time.

Description

Methods of Eliminating Intra-cell and Inter-cell Interference in TD-SCDMA System

[technical field]: the present invention belongs to wireless communication technical field, be specifically related to a kind of method that uses smart antenna (Smart Antenna) and joint detection (Joint Detection) technology to eliminate TD-SCDMA system inter-cell and intra-cell interference.

[Background Technology]: TD-SCDMA (Time Division-Synchronous Code Division Multiple Access) is my country's first third-generation mobile communication technology standard with independent intellectual property rights. Its proposal is of great significance to the independent innovation and industrial development of my country's communication industry.

The TD-SCDMA system adopts the time division duplex (TDD) mode. The reception and transmission are in different time slots of the same frequency, and different time slots are used to separate the reception and transmission channels. The basic principle is shown in Figure 2.

In a sense, TD-SCDMA is designed based on smart antenna technology. The smart antenna dynamically tracks the desired user's shaped beam to send and receive signals, greatly reducing system interference. Joint detection refers to the joint detection of a single user's signal by using information such as symbols, time, signal amplitude, and phase of multiple users to achieve better reception. Smart antennas can effectively suppress inter-cell interference; joint detection can effectively reduce intra-cell interference.

Inter-cell interference means that in a multi-cell system, each cell base station and user receive and receive signals relatively independently, and may interfere with each other. Interference between adjacent cells is mainly caused by synchronization deviation and different ratios of uplink and downlink time slot allocation. In addition, due to the different propagation delays of the cells that are far away, there will also be interference caused by time slot superposition between base stations.

The interference of intra-cell interference mainly includes intersymbol interference and multiple access interference. Intersymbol interference is for a single user signal. Due to the influence of the wireless propagation environment, the preceding and following codes or chips of the same user signal will be superimposed, making the receiver unable to resolve it smoothly, which results in inter-symbol interference (ISI). Intersymbol interference caused by multipath propagation cannot be completely eliminated in mobile communication systems. Like other CDMA systems, the TD-SCDMA system assigns mutually orthogonal address codes to modulate different user signals, but because only one same carrier frequency is used, interference occurs between user signals, which is called multiple access interference (MAI) .

[Summary of the invention]: The object of the present invention is to overcome the shortcomings of the prior art and provide a method for eliminating interference within and between cells in a TD-SCDMA system.

The method provided by the present invention for eliminating interference in a TD-SCDMA system cell and between cells is realized on the basis of a smart antenna in combination with joint detection technology, and the specific steps are as follows:

1. Eliminate inter-cell interference in TD-SCDMA system based on smart antenna technology

1.1. The composition of the smart antenna: including four parts: antenna array, analog-to-digital conversion, digital beamforming network and adaptive signal processor. The antenna array part adopts a uniform linear array with 8 array elements, and the distance between 2 array elements is λ/2; The schematic diagram of the connection relationship between the various parts of the smart antenna is shown in Figure 3, and the functions of each part are as follows:

Antenna array is a series of antenna array elements set separately in space, which can be divided into linear array, circular array and planar array according to different array elements. In practical applications, triangular arrays can also be formed according to different needs. Regular arrays and random arrays, etc., antennas of different arrays have different array responses to signals. The signal received by each array element is weighted, and the beam shape can be changed by changing the weight of the array. The antenna array is equivalent to discretely sampling the spatial signal in the airspace. Through signal processing, the beam generated by the array in the direction of the effective signal can be strengthened, and a "null" (Null) will be generated in the direction of the interference signal, thereby improving the system capacity. , Reduce system interference and expand system coverage.

Considering the smart antenna at the base station, the analog-to-digital converter converts the received analog signal into a digital signal during the uplink, and converts the processed digital signal into an analog signal during the downlink.

The main function of the beamforming network is that the antenna beam can adaptively adjust the weight coefficients w ₁ , w ₂ , ..., w _M through the digital signal processor according to the needs of the user and the transformation of the antenna propagation environment within a certain range , to tune to the appropriate beamforming network. Or select a group of optimal values from a pre-set weight coefficient list according to a certain criterion, so as to obtain the optimal direction of the main beam. The signal from the smart antenna forms a certain beam after passing through the beamforming network to realize the so-called spatial filtering. The beam after the beamforming network has a high gain in the direction of the desired signal, while the gain in the direction of the undesired interference signal is very low or zero, realizing the function of the notch.

Adaptive signal processing is an important aspect of smart antenna intelligence. It uses adaptive algorithms as the core to dynamically adjust the optimal weighting coefficients. The processing process is as follows:

1.2. The adaptive signal processor adjusts the weighted amplitude and phase of each array element through adaptive algorithms and optimization criteria, and dynamically generates spatial directional beams. The specific steps are:

1.2.1. First, calculate the output value of the beamformer based on the array signal and the current weight, and find the difference between the output of the beamformer and the expected signal; assuming that the signal received by the array antenna can be expressed as:

x(n)=[x ₁ (n), x ₂ (n), . . . , x _M (n)] ^H

The receiving weighting coefficient is:

w=[w ₁ , w ₂ , . . . w _M ] ^H

The output of the beamformer can be written as:

y(n)=w ^H (n)x(n)

The error of the shaper, that is, the difference between the output signal and the expected signal is:

e(n)=s(n)-y(n);

Section 1.2.2, automatically adjust the weighting weight of the beamformer according to the obtained difference

w(n+1)=w(n)+2μe(n)x(n)

Among them, μ is the step size factor, and the value of μe(n)x(n) affects the performance of the LMS algorithm. If the value of μe(n)x(n) is too small, the algorithm will converge slowly, and if the value is too large, the algorithm will be unstable. , and even diverge;

2. Eliminate interference in TD-SCDMA system cells based on joint detection technology

2.1. First, we assume that in the same frequency and the same time slot, there are K users transmitting data of limited length: the data sent by the kth user after QPSK modulation can be represented by vector d ^(k) ,

d(k)=(d ₁ ^(k) d ₂ ^(k) ,...d _N ^(k) ) ^T , k=1, 2,...K

The vector h ^(k) represents the channel impulse response of user k under the multipath Rayleigh fading channel, which can be accurately estimated by midamble in burst,

h(k)=(h ₁ ^(k) h ₂ ^(k) , . . . , h _W ^(k) ) ^T , k=1, 2, . . . K

The vector c ^(k) is the user's spreading code sequence, expressed as:

c ^(k) = (c ₁ ^(k) c ₂ ^(k) , . . . , c _Q ^(k) ) ^T , k=1, 2, . . . K

Among them, T represents the transpose, W is the effective sample value length of the sampling period _Tc of the impulse response (that is, the Chip period of the spreading code), and Q is the spreading factor, the maximum being 16, by the spreading sequence c ^(k) and h ^(k) vectors can get the combined channel impulse response b ^(k) of user k,

b ^(k) = c ^(k) *h ^(k)

With the above matrices and vectors, the total signal received by the receiver can be expressed as follows:

$\mathrm{<span\; class="notranslate">\; e</span>}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; K</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; e</span>}}^{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; K</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}\mathrm{<span\; class="notranslate">\; A</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; d</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; ad</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; no</span>}$

in,

e=(e ₁ , e ₂ , . . . , e _NQ+W-1 ) ^T , A=(A ⁽¹⁾ , A ⁽²⁾ , . . . , A ^(K) )

e is the total signal received by K users, and A is a matrix composed of combined channel impulse responses b ^(k) of K users. Quantization of the above formula can be obtained:

${\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; d</span>}}}_{\mathrm{<span\; class="notranslate">\; c</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; the\; lin</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; Me</span>}$

$<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; MAd</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; mn</span>}$

$<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; diag</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; MA</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; d</span>}<span\; class="notranslate">\; +</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; diag</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; MA</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; d</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; mn</span>}$

In the above formula, diag(X) represents the diagonal matrix of matrix X, that is, a matrix that only retains diagonal elements; and diag(X) represents a matrix that retains off-diagonal elements, that is, except that the diagonal elements are zero , and the remaining elements are reserved; thus, the first item diag(MA)d in the right side of the above formula diag(MA)d+diag(MA)d+Mn is a useful signal; the second item diag(MA)d is ISI and MAL; The three terms Mn are noise; the above formula clearly indicates the research direction of the linear joint detection algorithm, that is, the M matrix is selected according to the MMSE-BLE (minimum mean square error linear block equalizer) criterion, so that the last two terms in the above formula are MAI +ISI and noise have minimal impact on the estimates.

Advantage and positive effect of the present invention:

The method for eliminating the interference in the TD-SCDMA system sub-district and inter-subdistrict provided by the present invention is to combine the joint detection technology on the basis of the smart antenna in the TD-SCDMA system, so the present invention has the following advantages:

(1) Smart antennas eliminate inter-cell interference, and joint detection eliminates intra-cell interference, and the two need to be used together;

(2) The smart antenna alleviates the influence of the inaccurate channel estimation in the joint detection process on the deterioration of system performance;

(3) When the number of users increases, the calculation of joint detection is very large, and the use of smart antennas reduces the number of users;

(4) The number of array elements of the smart antenna is limited, and the smart antenna with M array elements can only suppress M-1 interference sources, and the formed side lobe is still interference to other users, so it can only be combined with joint detection to detect reduce these distractions;

(5) When the user moves at a high speed, the TDD mode adopts the same spatial parameters for the uplink and downlink, which makes the beamforming deviate; when the user is in the same direction, the smart antenna cannot play a role; there is also a problem caused by multipath with a delay exceeding one chip Intersymbol interference needs to be compensated by joint detection.

In a word, the combination of smart antenna and joint detection technology is extremely beneficial to the improvement of system performance. Joint detection can solve the interference that smart antennas cannot overcome when users are in the same direction; smart antennas can reduce the complexity of joint detection in multi-code channel processing, and completely eliminate the multiple access interference that joint detection cannot eliminate. Therefore, the combination of the two can reduce interference, thereby increasing the capacity of the system, obtaining high spectrum utilization, making the power control part in the base station and user equipment simpler, and reducing network costs at the same time.

[Description of drawings]:

Figure 1 is a flowchart of the simulation of the impact of smart antennas and joint detection on TD-SCDMA system interference.

Fig. 2 is a schematic diagram of the principle of the time division duplex (TDD) mode of the TD-SCDMA system involved in the present invention.

FIG. 3 is a schematic diagram of the principle structure of the smart antenna involved in the present invention.

Fig. 4 is a pattern diagram of a smart antenna according to a simulation result of the present invention.

FIG. 5 is a graph showing the relationship between the number of iterations and the error in the simulation process involved in the present invention.

FIG. 6 is a schematic diagram of the application structure of the joint detection technology involved in the present invention in the TD-SCDMA system.

Fig. 7 is a schematic diagram of the influence of the joint detection technology involved in the present invention on the interference in the cell of the TD-SCDMA system.

【Detailed ways】:

Example 1:

The method for eliminating interference within and between cells of a TD-SCDMA system provided by the present invention is realized on the basis of a smart antenna combined with joint detection technology, and the simulation flow chart is shown in FIG. 1 . In the simulation, the movement and switching of the user terminal are not considered, that is, the position of the user terminal is considered to be constant. The whole simulation process is mainly divided into the following three steps:

(1) Initial data preparation and setting, that is, generating data such as geography, propagation model, business model, base station and antenna;

(2) Simulation calculation, including multiple random processes and multiple iterative power control loops;

(3) Statistical mean value and data analysis.

Specific steps are as follows:

1. The method of eliminating inter-cell interference in TD-SCDMA system based on smart antenna technology:

(1) The basic idea of the smart antenna is to change the shape of the pattern of the entire array by changing the weighted amplitude and phase of each array element signal. It has the functions of direction finding and zero adjustment, that is, it can align the main beam with the incident of the desired user Signal and adaptive real-time tracking signal, while aligning the zero line or side lobe to the interference signal, thereby suppressing the interference signal, improving the signal-to-noise ratio of the signal, improving the performance of the entire communication system, and being able to identify direct waves in different incident directions in time and reflected waves.

The smart antenna is mainly composed of four parts: antenna array, analog-to-digital conversion, digital beamforming network and adaptive signal processor. The principle structure of the smart antenna is shown in Figure 2. Adaptive signal processing is an important aspect of the intelligent embodiment of the smart antenna. It takes the adaptive algorithm as the core and dynamically adjusts the optimal weighting coefficient.

(2), inter-cell interference

In a multi-cell system, each cell base station and users transmit and receive signals relatively independently, and may interfere with each other. Interference between adjacent cells is mainly caused by synchronization deviation and different ratios of uplink and downlink time slot allocation. In addition, due to the different propagation delays of the cells that are far away, there will also be interference caused by time slot superposition between base stations.

(3), simulation of interference elimination

The adaptive signal processor is the core of the smart antenna. The function of this unit is to adjust the weighted amplitude and phase of each array element according to certain algorithms and optimization criteria, and dynamically generate spatially directional beams. Spatial adaptive filtering (smart antenna) algorithms, including least mean square (LMS) algorithm, least squares (RLS) algorithm, sampling matrix inversion (SMI) and other algorithms.

The following is based on the least mean square (LMS) algorithm to simulate the effect of smart antennas on eliminating inter-cell interference.

①. The first step is to calculate the output value of the beamformer based on the array signal and the current weight, and calculate the difference between the output of the beamformer and the expected signal. Suppose the signal received by the array antenna can be expressed as:

x(n)=[x ₁ (n), x ₂ (n), . . . , x _M (n)] ^H

The receiving weighting coefficient is:

w=[w ₁ , w ₂ , . . . w _M ] ^H

The output of the beamformer can be written as:

y(n)=w ^H (n)x(n)

The error of the shaper, that is, the difference between the output signal and the expected signal is:

e(n)=s(n)-y(n)

②. In the second step, the weighting weight of the beamformer is automatically adjusted according to the calculated difference.

w(n+1)=w(n)+2μe(n)x(n)

where μ is the step size factor. The value of μe(n)x(n) affects the performance of the LMS algorithm. If the value is too small, the convergence speed of the algorithm is slow, and if the value is too large, the algorithm will be unstable or even diverge.

Simulation process:

The antenna array adopts a uniform linear array with 8 array elements, and the distance between 2 array elements is λ/2. The simulation results of using adaptive beamforming based on LMS to eliminate inter-cell interference are shown in Figure 3 and Figure 4.

Figure 3 is the radiation pattern of the antenna array. It can be seen that the smart antenna based on the adaptive beamforming LMS algorithm, the antenna array forms the maximum gain in the incident direction of the desired signal -20 degrees, and the interference signal direction -40 degrees, 0 degrees. The gain is almost 0, and the amplitude response in the main lobe direction is at least 10dB larger than other directions

Therefore, the smart antenna can form a main lobe in the desired signal direction and a deep null in the interference direction. Inter-cell interference has been suppressed, and the anti-interference ability of the system has been greatly improved.

Figure 4 shows the relationship between the number of iterations and the error. It can be seen that after 200 iterations, the LMS algorithm converges, and the error of the reference signal is controlled within a relatively small range;

2. The method of eliminating interference in the TD-SCDMA system cell based on joint detection technology:

(1) The basic idea of the joint detection technology is to use the prior information related to multiple access interference and combine the results of channel estimation to detect the signals of all users at the same time. Theoretically speaking, joint detection can completely offset the influence of multiple access interference and intersymbol interference, overcome the "near and far effect", and greatly improve the anti-interference ability of the system. However, in an actual system, considering the impact of realizability, especially the actual channel estimation, the impact of multiple access interference cannot be completely eliminated.

Joint detection (JD, joint detection) is the key technology of anti-jamming in CDMA mobile communication system. In the CDMA system, there is a certain correlation among the signals of various users, which is the root cause of the existence of multiple access interference. Although the MAI generated by individual users is very small, as the number of users or signal power increases, MAI becomes the main interference of wideband CDMA. The traditional detection technology is to perform spreading code matching processing on each user signal separately, so the anti-MAI interference ability is poor; the joint detection is based on the traditional detection technology, and the signals of multiple users are demodulated and output together. Make full use of useful information in MAI (spreading code sequence of each user, correlation between users and multipath, etc.), perform joint detection on each user or cancel mutual interference from the received signal, thereby effectively eliminating multiple access Interference and Intersymbol Interference.

(2), interference in the cell

Interference in a single carrier cell mainly includes intersymbol interference and multiple access interference. Intersymbol interference is for a single user signal. Due to the influence of the wireless propagation environment, the preceding and following codes or chips of the same user signal will be superimposed, making the receiver unable to resolve it smoothly, which results in inter-symbol interference (ISI). Intersymbol interference caused by multipath propagation cannot be completely eliminated in mobile communication systems.

Like other CDMA systems, the TD-SCCMA system assigns mutually orthogonal address codes to modulate different user signals, but because only one same carrier frequency is used, interference occurs between user signals, which is called multiple access interference (MAI) .

(3) First, we assume that in the same frequency and the same time slot, there are K users transmitting data of limited length: the QPSK-modulated data sent by the kth user can be represented by vector d ^(k) .

d(k)=(d ₁ ^(k) d ₂ ^(k) ,...d _N ^(k) ) ^T , k=1, 2,...K

Vector h ^(k) represents the channel impulse response of user k under multipath Rayleigh fading channel, which can be accurately estimated by midamble in burst.

h(k)=(h ₁ ^(k) h ₂ ^(k) , . . . , h _W ^(k) ) ^T , k=1, 2, . . . K

The vector c ^(k) is the user's spreading code sequence, expressed as:

c ^(k) = (c ₁ ^(k) c ₂ ^(k) , . . . , c _Q ^(k) ) ^T , k=1, 2, . . . K

Wherein, W is the effective sample length of the sampling period T _c of the impulse response (that is, the Chip period of the spreading code), and Q is the spreading factor, which is 16 at most. The combined channel impulse response b ^(k) of user k can be obtained from the product of spreading sequence c ^(k) and h ^(k) vectors.

b ^(k) = c ^(k) *h ^(k)

With the above matrices and vectors, the total signal received by the receiver can be expressed as follows:

$\mathrm{<span\; class="notranslate">\; e</span>}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; K</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; e</span>}}^{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; K</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}\mathrm{<span\; class="notranslate">\; A</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; d</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; ad</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; no</span>}$

in,

e=(e ₁ , e ₂ , . . . , e _NQ+W-1 ) ^T , A=(A ⁽¹⁾ , A ⁽²⁾ , . . . , A ^(K) )

e is the total signal received by K users, and A is a matrix composed of combined channel impulse responses b ^(k) of K users. Quantification of the above formula can be obtained:

$\begin{array}{c}{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; d</span>}}}_{\mathrm{<span\; class="notranslate">\; c</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; the\; lin</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; Me</span>}\\ <span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; MAd</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; mn</span>}\\ <span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; diag</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; MA</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; d</span>}<span\; class="notranslate">\; +</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; diag</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; MA</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; d</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; mn</span>}\end{array}$

In the above formula, diag(X) represents the diagonal matrix of the matrix X, that is, only the moments of the diagonal elements are retained; and diag(X) represents the matrix that retains the off-diagonal elements, that is, the diagonal elements are zero, the rest of the elements remain. In this way, the first item on the right side of the above formula is a useful signal; the second item is ISI and MAL; the third item is noise. The above formula clearly indicates the research direction of the linear joint detection algorithm, that is, the M matrix is selected according to the MMSE-BLE (Minimum Mean Square Error Linear Block Equalizer) criterion, so that the last two items in the above formula, namely MAI+ISI and noise pair Estimates have minimal impact.

From the uplink and downlink interference trend curve in Figure 7, it can be found that when the joint uplink and downlink detection is not used, the intra-cell interference (especially the uplink interference) increases significantly, and the inter-cell interference does not change significantly. It can be seen that the joint detection technology has a significant effect on eliminating the interference within the cell, and the suppression of the interference between the cells is better when the smart antenna is used.

Claims (1)

1. A method for eliminating interference between cells in TD-SCDMA system is characterized in that the method is realized by combining a joint detection technology on the basis of an intelligent antenna, and comprises the following specific steps:

1, eliminating inter-cell interference of TD-SCDMA system based on intelligent antenna technology

1.1, composition of the intelligent antenna: the system comprises an antenna array part, an analog-to-digital conversion part, a digital beam forming network part and a self-adaptive signal processor, wherein the antenna array part adopts a uniform linear array of 8 array elements, and the distance between 2 array elements is lambda/2;

1.2, the adaptive signal processor adjusts the weighted amplitude and phase of each array element through an adaptive algorithm and an optimization criterion to dynamically generate a spatial directional beam, and the method comprises the following specific steps:

1.2.1, firstly, calculating the output value of a beam former according to the array signal and the current weight, and calculating the difference value between the output of the former and the expected signal; assume that the signal received by the array antenna can be expressed as:

x(n)＝[x₁(n)，x₂(n)，...，x_M(n)]^H

the receive weighting coefficients are:

w＝[w₁，w₂，...w_M]^H

the output of the beamformer can be written as:

y(n)＝w^H(n)x(n)

the error of the shaper, i.e. the difference between the output signal and the desired signal, is:

e(n)＝s(n)-y(n)；

wherein: h represents conjugate transpose, and M represents antenna array element number;

1.2.2, the weighting weight of the beam former is automatically adjusted according to the obtained difference

w(n+1)＝w(n)+2μe(n)x(n)

Wherein μ is a step factor, the value of μ e (n) x (n) affects the performance of the LMS algorithm, the convergence rate of μ e (n) x (n) is slow when the value is too small, and the algorithm is unstable and even diverges when the value is too large;

eliminating interference in TD-SCDMA system cell based on joint detection technology

2.1, first, we assume that in the same frequency and the same time slot, there are K users transmitting data of limited length: wherein the usable vector d of QPSK modulated data sent by the k-th user^(k)It is shown that,

d(k)＝(d₁ ^(k)d₂ ^(k)，...d_N ^(k))^T，k＝1，2，...K

vector h^(k)Represents the impulse response of the k channel of the user under the multipath Rayleigh fading channel and can pass through the burstTo accurately estimate the midamble of (a) the mobile station,

h(k)＝(h₁ ^(k)h₂ ^(k)，...，h_W ^(k))^T，k＝1，2，...K

vector c^(k)The spreading code sequence for a user is expressed as:

c^(k)＝(c₁ ^(k)c₂ ^(k)，...，c_Q ^(k))^T，k＝1，2，...K

where T represents transpose and W is the sampling period T of impulse response_cI.e. the effective sample length of the Chip period of the spreading code, Q is the spreading factor, max 16, from the spreading sequence c^(k)And h^(k)The product of the vectors can obtain the combined channel impulse response b of the user k^(k)，

b^(k)＝c^(k)*h^(k)

With some of the above matrices and vectors, the total signal received by the receiver can be represented as follows:

<math> <mrow> <mi>e</mi> <mo>=</mo> <munderover> <mi>&Sigma;</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>K</mi> </munderover> <msup> <mi>e</mi> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> </msup> <mo>+</mo> <mi>n</mi> <mo>=</mo> <munderover> <mi>&Sigma;</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>K</mi> </munderover> <mi>A</mi> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> <mi>d</mi> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> <mo>+</mo> <mi>n</mi> <mo>=</mo> <mi>Ad</mi> <mo>+</mo> <mi>n</mi> </mrow></math>

wherein,

e＝(e₁，e₂，...，e_NQ+W-1)^T，A＝(A⁽¹⁾，A⁽²⁾，...，A^(K))

e is the total signal received by K users, A is the combined channel impulse response b of K users^(k)The matrix formed by the method is obtained by quantizing the above formula:

${\hat{d}}_{c,\mathrm{lin}}=\mathrm{Me}$

$=\mathrm{MAd}+\mathrm{Mn}$

<math> <mrow> <mo>=</mo> <mi>diag</mi> <mrow> <mo>(</mo> <mi>MA</mi> <mo>)</mo> </mrow> <mi>d</mi> <mo>+</mo> <mover> <mi>diag</mi> <mo>&OverBar;</mo> </mover> <mrow> <mo>(</mo> <mi>MA</mi> <mo>)</mo> </mrow> <mi>d</mi> <mo>+</mo> <mi>Mn</mi> </mrow></math>

in the above formula, diag (X) represents a diagonal matrix of matrix X, that is, a matrix with only diagonal elements reserved; diag (x) denotes a matrix that retains off-diagonal elements, i.e., the remaining elements are retained except for the diagonal elements, which are zero; thus, in the above formula, the 1 st diag (MA) d of the right-hand diag (MA) d + Mn is the useful signal; item 2 diag (ma) d ISI and MAL; the third term Mn is noise; the above equation clearly indicates the research direction of the linear joint detection algorithm, i.e. MMSE-BLE, i.e. the criterion of minimum mean square error linear block equalizer selects M matrix, so that the last 2 terms in the above equation, i.e. MAI + ISI and noise, have as little influence on the estimated value as possible.

Images