CN1352827A - Signal identification in CDMA ratio systems - Google Patents

Abstract

According to the inventive method for identifying signals in radiocommunications systems, subscriber-specific narrow-band pilot signals with a predetermined power are transmitted in addition to the bursts that are transmitted. At the receiving end, characteristic values of the pilot signals transmitted are obtained from the mixture of different subscriber signals received. These characteristic values are used to identify the origin of the incoming subscriber signals and/or for determining their received power and/or for determining their Doppler displacements and/or for obtaining other information.

Description

Signal Identification in Code Division Multiple Access Wireless Systems

The invention relates to a method and device for signal identification in a CDMA wireless system.

An important design part of the cellular mobile radio system lies in the realization of the radio interface. A typical radio interface scheme is TD-CDMA (Time Division-Code Division Multiple Access), see DE 198 22 276. It is characterized in that the interference in the cell, ie the interference originating in the reference cell under consideration, is eliminated in the receiver of the reference cell by means of joint detection (cf. DE 41 21 356). As a capacity-limiting factor there is also inter-cell interference from other cells, which cannot be eliminated in the reference cell with the joint detection process according to the prior art. Provided that at least the most intense sources of inter-cell interference incident are successfully eliminated within said reference cell, a huge increase in capacity can be achieved.

The use of joint detection to eliminate interference in the cell is based on the following content, that is, at the receiver side

- estimating the channel impulse response of the reference cell in a first step from the transmitted known midamble signal, and

- Estimating the user's data in a second step from said estimated channel impulse response and the known CDMA code of the user in the reference cell.

Both estimation procedures described above are based on solving a system of equations, where the number of unknowns, ie channel coefficients and data symbols, should not be greater than the number of equations available. This condition can be complied with by suitable selection of the mid-sequence length _Lm , the number of simultaneously active users K and the length Q of the CDMA code.

At present, it has not been possible to successfully incorporate the inter-cell interference into the interference cancellation process performed in the reference cell because of the large number of Potential sources of inter-cell interference. Such potential sources of inter-cell interference are all transmitters of the adjacent cell which use the same frequency band and - TDMA frame - time slot as the subscribers of the reference cell under consideration. If all these sources of interference are considered in the cancellation process of the reference cell, this process requires channel estimation and data estimation for the user signal incident from the external cell into the receiver of the reference cell, which leads to the equation The number of equations in the system is less than the number of unknowns, so no solution can be obtained.

EP 96 118 916 has disclosed that only a part of the potential inter-cell interference sources that produce effective interference power to the receiver of the reference cell are always considered. The current interference effect depends on the position of the interference source and the shadowing effect. If it is limited to eliminating those inter-cell interference sources with relatively high power, it can be realized that when channel estimation and data estimation are performed in the reference cell, the number of unknowns will not exceed the number of available equations. However, the precondition for this limitation is that it must be known in the receiver of the reference cell which of the multiple potential sources of inter-cell interference currently have a higher incident power and is therefore very critical . If the critical inter-cell interference source is known, it is also necessary to know its midamble code and CDMA code. After obtaining these basic knowledge, the interference source can be included in the interference elimination processing of the reference cell.

The object of the invention is to identify the critical source of inter-cell interference in the reference cell and, if necessary, to obtain additional information on this source of interference. This object is achieved by a method with the features of claim 1 and an apparatus with the features of claim 14 . Preferred developments of the invention are given by the subclaims.

In the signal identification of the wireless communication system, in addition to the user signal sent, the present invention also sends some user-specific narrowband pilot signals with predetermined power. On the receiving side, the feature quantity of the sent pilot signal is obtained from the received mixed signal of different user signals. The feature quantity is used to identify the source of the incident user signal and/or to determine its received power and/or to determine its Doppler shift and/or to obtain other information.

Using this method, some user signals can be pre-selected, and these signals can greatly reduce the cost of subsequent channel estimation and detection. Inter-cell interference in the detection can thus be taken into account with suitable signal processing outlay. Thereby improving the system capacity.

Exemplary embodiments of the invention are explained in more detail below with reference to the drawings. in:

Figure 1 shows two different signals consisting of an intermediate sequence and a data-carrying part,

Figure 2 shows a diagram of the signal processing of two signals,

Fig. 3 shows a diagram of signal processing of a large signal group.

Figures 1 and 2 show an embodiment based on the idea of the present invention with an example of a simple wireless communication system with a TD-CDMA user separation method, wherein said wireless communication system has only two potential inter-cell interference sources 1, 2 . Figure 1a shows the radio block (burst) of the useful transmitted signal i ₁ (t) with bandwidth B consisting of two data-carrying parts and an intermediate sequence, and also shows the relationship with the burst i ₁ ( t) The pilot signal p ₁ (t) of the superimposed inter-cell interference source 1, wherein the burst time delay is represented by T _bu .

Figure 1b shows a corresponding embodiment of an inter-cell interferer 2 . In this embodiment, the pilot signals p ₁ (t) and p ₂ (t) are sinusoidal signals with different frequencies. The frequencies of the sinusoidal signals should be so far apart that they can be distinguished despite a possible Doppler shift. It is assumed here that the pilot signal has a constant and known transmit power. However, it is sufficient that the transmission power of the pilot signal can be known at the receiving side, for example, by means of signaling. That is, the parameters of the pilot signal do not have to remain constant in order to be able to match the changing conditions of the radio interface. In order to replace the sinusoidal signal, any complex-valued signal can also be used as the pilot signal, such as a channel test sequence, a known chip sequence, a PN sequence (binary signal), and the like.

The following signal is received in the receiver of the reference cell:

s(t): The sum of the desired received signals from the reference cell with a bandwidth of B

a ₁ [i ₁ (t)+p ₁ (t)]: inter-cell interference source 1 attenuated with transmission factor 0<a ₁ <1

transmit a signal

a ₂ [i ₂ (t)+p ₂ (t)]: inter-cell interference source 2 attenuated with transmission factor 0<a ₂ <1

transmit a signal

The total signal incident on the receiver of the reference cell is then given by:

e(t)=s(t)+a ₁ [i ₁ (t)+p ₁ (t)]+a ₂ [i ₂ (t)+p ₂ (t)] (1).

FIG. 2 shows a device for obtaining the characteristic quantities of the pilot signals p ₁ (t) and p ₂ (t). In this device, received signals e(t) as represented by (1) are input to a filter for signal matching with p ₁ (t) and p ₂ (t), respectively. Without loss of generality, one can set: $\underset{0}{\overset{{\mathrm{<figure-callout\; id="1"\; label="T\; bu\; p"\; filenames="A0080794400091.png"\; state="\{\{state\}\}">T</figure-callout>}}_{\mathrm{bu}}}{\&Integral;}}{p}_{}^{}{}_1^2\left(t\right)\mathrm{dt}=\underset{0}{\overset{T_{\mathrm{bu}}}{\&Integral;}}{p}_2^2\left(t\right)\mathrm{dt}=1---\left(2\right)$

If the filter output signal is sampled at the end of the burst, i.e. at the time point t=T _bu , two test variables are obtained consisting of the useful term a ₁ , a ₂ and the disturbing term n ₁ or n ₂ . ${\mathrm{<figure-callout\; id="1"\; label="T"\; filenames="A0080794400091.png"\; state="\{\{state\}\}">T</figure-callout>}}_1=\underset{0}{\overset{T_{\mathrm{bu}}}{\&Integral;}}e\left(t\right)p_1\left(t\right)\mathrm{dt}=a_1+{\mathrm{no}}_1---\left(3\right)$ $T_2\underset{0}{\overset{T_{\mathrm{bu}}}{\&Integral;}}e\left(t\right)p_2\left(t\right)\mathrm{dt}=a_2+{\mathrm{no}}_2---\left(4\right)$

When seeking this test variable, the process gain is the product of the bandwidth B and the burst time delay T _bu . Therefore, using the power N ₁ of the signal e(t)-a ₁ p ₁ (t) and the power N ₂ of the signal e(t)-a ₂ p ₂ (t), the variance of n ₁ and n ₂ is approximately $\mathrm{E.}\left\{{\mathrm{no}}_1^2\right\}=\frac{{\mathrm{<figure-callout\; id="1"\; label="N"\; filenames="A0080794400091.png"\; state="\{\{state\}\}">N</figure-callout>}}_1}{{\mathrm{BT}}_{\mathrm{bu}}}---\left(5\right)$ $\mathrm{E.}\left\{{\mathrm{no}}_2^2\right\}=\frac{N_2}{{\mathrm{BT}}_{\mathrm{bu}}}---\left(6\right)$

If for example parameters B=5MHz and _Tbu =625μs as suggested by ETSI for UMTS, the process gain

_BTbu =3125≌35dB (7).

Such a high process gain means that only an extremely weak pilot signal is required to obtain sufficiently reliable test variables T ₁ and T ₂ compared to the useful signal. Thus, T ₁ and T ₂ are reliable measures of the strength of the two inter-cell interference signals i ₁ (t) and i ₂ (t) incident on the reference cell and are consistent with the characteristic quantities.

In addition, two threshold value discriminators with a threshold value of λ are installed in the device shown in FIG. 2 . If _T1 is greater than λ, the inter-cell interferer 1 is considered critical and included in the cancellation process in the reference cell, otherwise not. This also applies correspondingly to the inter-cell interferer 2 . The two threshold discriminators enable identification of the source of the signal.

In the case that K ₁ is greater than two potential sources of inter-cell interference, the solutions shown in FIGS. 1 and 2 can be extended according to FIG. 3 . If the frequency f _μ of K ₁ pilot signals pμ (μ=1…K ₁ ) is selected, the difference between f _μ and f _ν will satisfy the following conditions:

|f _μ -f _ν | T _bu ∈ N, μ, ν ∈ {1, 2, ..., K ₁ }, (8)

In this way, the pilot signals do not interfere with each other when obtaining the feature quantities.

Next, set forth in detail some other expansion schemes of the present invention:

In the method shown in Figures 1 and 2, the frequency selectivity of the mobile radio channel is not taken into account. With this frequency selectivity, the pilot signal will not be distorted due to its narrow-band nature, but may in some cases be so strongly attenuated that critical sources of inter-cell interference can no longer be identified as such. a major source of interference. This problem can be solved by using a user-specific combination of several different sinusoidal signals in the form of pilot signals for each subscriber station. It can thereby be ensured that the received overall pilot power will depend less on the instantaneous frequency characteristics of the radio channel due to frequency diversity, but only on the intensity of the relevant inter-cell interference sources incident on the reference cell .

In the solutions shown in Figures 1 and 2, the precondition is that the incident time point of the burst of the potential inter-cell interference source is known in the receiver of the reference cell, that is, the transmitter Synchronization is performed across cell boundaries. If the pilot signal is not only transmitted during the burst of the corresponding inter-cell interferer, but also outside the burst or continuously by the inter-cell interferer, the said inter-cell interferer can also be canceled. Synchronize. Some other extensions stipulate that the pilot signal

- transmit only within a fraction of the time delay of said transmitted user signal (less interference),

- transmitted within the entire time delay of said transmitted user signal,

- also transmitting in a time zone outside the time delay of said transmitted user signal,

- transmit in all or part of all radio blocks of the corresponding user signal (higher test accuracy),

- Launch with time breaks.

The pilot signal shows a certain interference to the useful signal, but on the other hand, the pilot signal is also interfered by the useful signal. It is therefore more expedient to separate the frequency spectrum of the pilot signal from the associated useful signal by, for example, setting a frequency range which is exclusively used by the pilot signal.

In the case of using an unmodulated sinusoidal signal as a pilot signal, if Fourier transform is performed on the total signal incident on the receiver of the reference cell, and the observation interval is a burst time delay T _bu , Indicative signals of all significant sources of inter-cell interference can then be shown in the frequency spectrum with isolated spectral lines. From the height and position of these spectral lines it can be concluded which potential sources of inter-cell interference are important. From the positions of these spectral lines, information about the Doppler shift of interfering signals between cells can also be seen, and thus the direction of motion of the subscriber station can be obtained. Alternatively, in order to obtain the characteristic amount of the transmitted pilot signal, a frequency estimation algorithm such as Esprit or Music may also be used.

In the schemes shown in Figures 1 and 2, the precondition is that the criticality of the source of inter-cell interference is determined after a single received burst. The reliability of this determination can be improved by taking into account the test variables obtained on the basis of the pulse train as intermediate variables and then determining the characteristic variable sought from N such intermediate variables. This scheme is shown in Figure 3.

The invention can also be applied to base stations or subscriber stations using antenna groups with a plurality of antenna elements. Different pilot signals can be transmitted through different antenna units, or pilot signals can only be applied to a part of the antenna units. For the receiving side, the use of antenna groups means that the received signals of the individual antenna elements can be analyzed according to different algorithms.

In addition to the user signal, the transmitting side also transmits a narrow-band pilot signal, while the receiving side analyzes the pilot signal before data detection. Through the cooperation between the transmitting and receiving sides, it is possible to obtain a method that only improves the detection quality. A device that eliminates significant sources of interference (wireless communication systems).

If an important user signal has been selected with the aid of the analyzed pilot signal, then the channel estimation and data detection described can be simplified by only considering this important user signal, see EP 96 118 916 for this. Then, for example, the JD method (JD joint detection) is used for signal analysis, and the JD method can calculate the influence of each interference on the user signal to be analyzed in the form of joint detection.

Claims (14)

1. A method of identifying a signal in a wireless communication system, wherein, The base station and the subscriber station are connected through a wireless interface, and transmit user signals through the wireless interface, which is characterized in that: - sending some user-specific narrowband pilot signals in addition to said transmitted user signals, - at the receiving side, obtaining the characteristic quantities of said transmitted pilot signal from the received mixed signal, - the characteristic quantity is used to identify the source of the user signal incident on the receiver and/or to determine its received power and/or to determine its Doppler shift and/or to for additional information. 2. The method of claim 1, wherein: The narrowband pilot signal is sent with a predetermined power known at the receiving side. 3. The method according to any one of the preceding claims, characterized in that: The wireless interface is constructed according to the TD-CDMA user separation method. 4. The method according to any one of the preceding claims, characterized in that: The base station uses an antenna group with multiple antenna elements, transmitting different pilot signals through said different antennas, and/or not applying pilot signals to all antenna elements, On the receiving side, the received signals of the individual antenna elements are analyzed according to different algorithms. 5. The method according to any one of the preceding claims, characterized in that: Said pilot signal is an unmodulated sinusoidal signal having a frequency characteristic of the corresponding subscriber station. 6. The method according to any one of the preceding claims, characterized in that: The pilot signal is composed of a plurality of unmodulated sinusoidal signals with different frequencies and the same or different power, the frequency or frequency combination of which characterizes the corresponding subscriber station. 7. The method according to any one of the preceding claims, characterized in that: The pilot signal is a narrow-band modulated signal, wherein analog or digital methods such as amplitude modulation, phase modulation and frequency modulation are used for modulation, or a combination of these methods. 8. The method according to any one of the preceding claims, characterized in that: Said pilot signals are transmitted in sub-regions of the frequency range required by the corresponding user signals. 9. The method according to any one of claims 1 to 7, characterized in that: Said pilot signals only or also have spectral components which lie outside the desired frequency range of the corresponding user signal. 10. The method according to any one of the preceding claims, characterized in that: In order to obtain the characteristic quantities of the transmitted pilot signal from the received mixed signal, filters or correlators are used. 11. The method according to any one of the preceding claims, characterized in that: In order to obtain the characteristic quantities of the transmitted pilot signal from the received mixed signal, a Fourier transformation is applied. 12. The method according to any one of the preceding claims, characterized in that: To obtain the characteristic variables, intermediate variables are first ascertained by using short time segments of the received signal, from which the desired characteristic variable is then determined. 13. The method according to any one of the preceding claims, characterized in that: The characteristics of the pilot signal, such as frequency, spectrum, modulation, etc., vary with time. 14. Apparatus for performing the method of any one of the preceding claims.

Images