WO2003017510A1

WO2003017510A1 - Sinr estimation method and apparatus therefore

Info

Publication number: WO2003017510A1
Application number: PCT/CN2001/001226
Authority: WO
Inventors: Shaojun Xu; Yan Gao
Original assignee: Linkair Communications, Inc.
Priority date: 2001-08-16
Filing date: 2001-08-16
Publication date: 2003-02-27
Also published as: CN1475049A

Abstract

A SINR estimation method and apparatus in a communications system are disclosed. The SINR estimation method comprises the steps of converting an I component and a Q component of the received interfered signals into decided symbol; mapping the decided symbol into the QAM constellation point to obtain the estimated signal; delaying the I component and the Q component of the received interfered signal; subtracting the estimated signal from the received interfered signal to obtain the estimated interference; calculating the power of the estimated interference and the power of the estimated signal; and determining the SINR based on the power of the estimated interference the power of the estimated signals. The SINR estimation method and the apparatus of the present invention can quickly and accurately estimated the SINR ratio, which solves the defects, such as low computing speed, high complexity.

Description

SINR ESTIMATION METHOD AND APPARATUS

THEREFORE

FIELD OF THE TNVENTTON

The invention relates generally to an apparatus and method for measuring Signal to Interference plus Noise Ratio (SINR) in a communications system. More particularly, the present invention relates to a novel and improved technique of SINR measurement method for measuring SINR, which is the ratio of a desired signal power to interference plus noise power.

BACKGROUND OF THE INVENTION

In order to control and monitor communication quality in wireless communication, it is required that the signal-to-interference plus noise ratio of a received signal be measured in the receiver. Especially, in modern cellular mobile communication, accurate SINR should be estimated in a lot of algorithms such as power control, AMC (Adaptive modulation coding) and Turbo coding etc. A lot of researches in this field have been done to require accurate SINR.

For example: M. D. Austin and G. L. Stuber proposed a SINR estimation method by using a training sequence. The estimation accuracy depends on symbol error characteristic.(shown in article "In service signal quality estimation for

TDMA cellular systems " in Proc.PIMRC,1995,pp.836-840)

Andersin in his article "Subspace based estimation of the signal to interference ratio for TDMA cellular systems"(in Proc. VTC, Atlanta, GA, 1996, pp.1155-1159), proposed another method that SINR can be achieved by analyzing the variance matrix characteristic root of the received signal.

In 1998 Turkboylarii and G. L. Stuber proposed a SINR estimation method based on sub-space in the article "An efficient algorithm for estimating the signal-to-interference ratio in TDMA cellular systems" ( IEEE Trans. Commun.,

vol.46,ρp. 728-731, June 1998 ) . But its estimation time is too long for rapid power control, and its high complexity is a fatal limitation.

The simulation results show that the last two methods have a better performance than the first method. These methods have too high complexity and need a long sequence, which results in their limitation in practical application such as fast power control.

WCDMA proposed a SINR estimation method by computing the variance of received signals. But it needs an important condition that the mean of interference is equal to zero. In WCDMA and IS-2000, PN code spreading is used to overcome it. But this method has several limitations. Firstly the receiver has a high complexity because of MUD (Multi-User Detection). Secondly the pilot channel estimation is added to redundancy and results in capacity loss. Thus this method has a limitation in practical application. Unfortunately, however, conventional SINR estimation methods mentioned above tend to low speed and high complexity, such as complicated circuit arrangement and complicated arithmetic operations.

SUMMARY OF THE TNVENTTON Accordingly, it is an object of the present invention to provide an improved method to estimate the Signal-Interference-Pius-Noise-Ratio of the wireless link rapidly and accurately.

Accordingly, another object of the present invention is to provide a SINR measurement apparatus in which SINR can be measured highly accurately and through a simple arrangement and simple operations.

In accordance with the present invention, the foregoing objects is attained by providing a SINR measurement apparatus in a communications system, comprising: a demodulator for converting an I component and Q component of the received interfered signals into decided symbol; a modulator, coupled to the demodulator, for mapping the decided symbol into the QAM constellation point to obtain the estimated signal; delay means for delaying the I component and the Q component of the received interfered signal; subtraction means for subtracting the estimated signal from the received interfered signal to obtain the estimated interference; calculation means, coupled to the modulator and the subtraction means, for calculating the power of the estimated interference and the power of the estimated signal; and determination means for determining the SINR based on the power of the estimated interference the power of the estimated signals.

Further, the foregoing object of the present invention is achieved by providing an providing a SINR measurement method, comprising the steps of : converting an I component and a Q component of the received interfered signals into decided symbol; mapping the decided symbol into the QAM constellation point to obtain the estimated signal; delaying the I component and the Q component of the received interfered signal; subtracting the estimated signal from the received interfered signal to obtain the estimated interference; calculating the power of the estimated interference and the power of the estimated signal; and determining the SINR based on the power of the estimated interference the power of the estimated signals.

In accordance with the present invention, the accuracy of SINR Estimation is higher when the average signal-to-interference is high. Further, within a cellular system, when the receiver is nearby the transmitter, the accuracy of SINR Estimation is high as the channel estimation is more reliable and the interference coming from other adjacent cell is also small. In addition, the accuracy of estimation can become higher when the number of time slots which will be estimated increasing.

The method described as follows are based on a 16QAM scheme communication system, other modulation scheme can also adopt this method.

BRTEFE DESCRIPTIONS OF THE DRAWINGS The accompanying drawings which are incorporated in and constitute a part of this specification, illustrate particular embodiments of the invention, and together with the description, serve to explain, and not restrict, the principles and advantages of the present invention.

Figure 1 shows the principle of Constellation Decision Feedback SINR estimation of present invention;

Figure 2 is a block diagram showing the arrangement of a SINR measurement apparatus according to the first embodiment of the present invention;

Figure 3 is a block diagram showing the arrangement of another SINR measurement apparatus according to the second embodiment of the present invention;

Figure 4 is a chart showing the SINR estimation performance in the transmission scheme;

Figure 5 is a chart showing another SINR estimation performance in the transmission scheme; and Figure 6 is a topology of a 7-cellular structure.

DETAILED DESCRIPTIONS OF THE INVENTION

As we have known, in a wireless communication system, a transmitter transmits serial data alternately one bit at a time to split the data into two sequences, namely in-phase component (I-component) data and quadrature component (Q-component) data. The data in each of the two sequences is spread- spectrum modulated by spread spectrum code. QPSK (quadrature phase-shift keying) modulation is usually applied to spread-spectrum modulated signals of the I and Q components and the resulting signal is transmitted. In the receiver, it has an antenna, a filter which only passes the necessary frequency band, and a quadrature demodulator which demodulates spread-spectrum signals of the I and Q components.

Figure 1 shows the principle of Constellation Decision Feedback SINR estimation of present invention.

With reference to Figure 1, we assume that the I component and the Q component of the modulated signal at the transmitter is (χ,y) . The I component

and the Q component of despread-spectrum signals, which is the actual received value at receiver, is (x',y') . The I component and the Q component at the output of

the decider at the receiver is (χ,y) . The above-mentioned coordinates are shown

within the constellation. From Figure 1 , it is easy to see the relation between the signal vector and interference vector.

The ideal signal and interference power are:

S = (x² + y²) ,

- χ'γ ₊ (y ~_y ^>γ (1)

where S is the ideal signal power, I is the ideal interference power , x is the I component of the modulated signal at the transmitter; y is the Q component of the modulated signal at the transmitter, x^' is the I component of despread-spectrum signals at the receiver; y^' is the Q component of despread-spectrum signals at the receiver. The estimated signal and interference power are: S = (xxf) , ϊ = (x - x 'Y + (y - y ')² (2)

where S is the estimated signal power, / is the estimated interference power , x

is the I component at the output of the decider; j> is the Q component at the output

of the decider; x^' is the I component of despread-spectrum signals at the receiver; y is the Q component of despread-spectrum signals at the receiver. If the output of the decider is correct, then:

S = S,/ = / (3)

The relative accurate value can be obtained within a period of time. Thus, the signal to interference plus noise ratio (SINR) can be obtained from formula (4):

where T is the time slot accumulation number. It is determined by measurement period. S_m is the estimated signal power at the m-th symbol. ϊ_m is the estimated

interference power at the m-th symbol.

Fig. 2 is a block diagram showing the arrangement of a SINR measurement apparatus according to the first embodiment of the present invention.

As shown in Fig. 2, the SINR measurement apparatus comprises a demodulator 201, a modulator 204, delay means 202, 203, subtraction means 205, 206, calculate means 207, 208, accumulation average means 209, 210 and determination means 211.

With reference to Figure 2, the I component and the Q component of despread- spectrum signals (χ',y'), which is the actual received value at receiver, is inputted

into the demodulator 201. Then, the demodulator 201 converts the I component and Q component of the received interfered signals, into decided symbol separately and output to the modulator 204. After having received the decided symbol from the demodulator 201, the modulator 204 maps the decided symbol into the QAM constellation point to obtain the estimated signal (x, j>) . In Figure 2, a pair of delay

means 202 and 203 are used to delay the I component and the Q component of the received interfered signal. In the present embodiment, the delay time of delay means 201, 201 is equal to the time that received interfered signals passing through the demodulator 201 and the modulator 204. Referring to Figure 2, the delayed I and Q component of the received interfered signals (χ',y') and the estimated signals

( , j>) from the modulator 204 are subtracted in the subtraction means 205 and 206

to obtain the estimated interference x-x', y-y' . The calculation means 207, 208, which connects to the modulator 204 and subtraction means 205, 206, calculates the power of the estimated interference and the power of the estimated signal respectively. The accumulation average means 209, 210 determines the accumulated values of the estimated signals and estimated interference separately. Then, based on the estimated signals and the estimated interference, the determination means 211 can determine the SINR from formula (4).

In the present invention, the signal can be separated from the interfered signal. The difference between the signal and the interfered signal is the interference. Then the SINR can determined by the SINR estimation apparatus. Further, with reference to Figure 2, the present embodiment will be described in detailed by using a m-th example.

At the m-th sample, the input I and Q components of the demodulator 201 are x_m' , y_m respectively. That is, the received interfered signal of demodulator 201 is

(x_m' ,y_m' )• The interfered signal

) is the output of despreader in the receiver. The output of demodulator 201 is the decided symbol. The decided symbol is inputted into the modulator 204 and then is mapped into the constellation point The output of modulator 204 is estimated signal (x_m,y_m ), and the estimated

signal power is:

The interfered signal (x_m' ,y_m' ) also passes through the delay means 202, 203. We assume that time Δ is the delay of the demodulator 201 and the modulator 202. The interfered signal (x_m' ,y_m' ) and the estimated signal {x_m,y_m ) are inputted into the subtraction means 205, 206, thus, the output of which is the estimated interference x' - x , y ' — γ ).

The estimated interference power is l =« - +(y_m' -y (6)

The calculate means 207, 208 are used to compute the power of the estimated interference I_m and the power of the estimated signal S_m respectively.

The estimated signal power S„, and the estimated interference power 7_mare separately inputted into the accumulation average means 209, 210. We assume that T is the accumulation time slots, therefore the output of accumulation average

means 209, 210 are TS,-_. ^and __I_m respectively. Then the determination means m=0 "^,=0

211 can determine that the estimated the SINR in dB is:

Fig 3 is a block diagram showing the arrangement of another SINR measurement apparatus according to the second embodiment of the present invention.

In this embodiment, if the constellation power is known in advance, the step of estimating signal is unnecessary. Otherwise, the estimation error will be introduced.

Referring to Figure 3, the present embodiment will be also described in detailed by using a m-th example.

At the m-th sample, the input I and Q components of demodulator 301 are _> y_m' respectively. The received interfered signal of demodulator 301 is

\x_m' ,y_m' ). The interfered signal \x_m' ,y_m' ) is the output of despreader in the receiver.

The output of demodulator 301 is the decided symbol. The decided symbol is inputted into the modulator 304 and then is mapped into the constellation point

(x , y ) ■

The output of modulator 304 is estimated signal (x_m,y_m ), and the estimated

signal power is:

The interfered signal (x_m' ,y„', ) also passes through the delay means 302, 303.

We assume that time Δ is the delay of the demodulator 301 and the modulator 302. The interfered signal x_m' ,y_m' ) and the estimated signal {x_{m >}y_m ) are inputted into the subtraction means 305, 306, thus, the output of which is the estimated interference x' - x„, V - v ).

The estimated interference power is

The calculate means 307 is used to compute the power of the estimated interference I_m .

The estimated interference I_m is inputted into the accumulation average means 309. We assume that T is the accumulation time slots, therefore the output of accumulation average means 309 is __I_m . m=0 In the present embodiment, the parts for estimating the signal power shown in Fig. 2 is replaced by the constellation power determination means 308. In present SINR method, since the constellation power is known beforehand, we only need estimate the interference. The method of estimating interference power is same as that of the first embodiment. Then, the determination means 310 can determine that the estimated the SINR in dB is:

where S_comlellalton the constellation power, T is the accumulative time slot

number. ϊ_m is the power of the estimated interference..

From above description, it can be seen that this method can quickly and accurately estimate the Signal-Interference-Ratio of the desired channel.

Figure 4 is a chart showing the SINR estimation performance in the transmission scheme. Figure 6 is a topology of a 7-cellular structure. Referring to figure 4, this scheme adopts a 20ms frame structure, each frame consists of 160 timeslots. The topology of the 7-cellular structure is illustrated in figure 6. Figure 4 shows the relationship between the SINR estimation error variance σ² , which is the vertical coordinate of Fig.4, and the normalized distance from mobile station to base station along the cell's radius, the estimation error variance σ² is in unit of dB.

Figure 5 is a chart showing another SINR estimation performance in the transmission scheme.

Referring to figure 5, this scheme also adopts 20ms frame structure, each frame consists of 160 timeslots. The topology of the 7-cellular structure is illustrated in figure 6. Figure 5 illustrates the relationship between the SINR estimation error power Δ², which is the vertical coordinate, and the normalized distance from mobile station to base station along the cell's radius, the estimation error power Δ² is in unit of dB.

The definition of error variance σ² :

The definition of error power Δ²

Δ^z *fe-gri

E[(SIR - SIR)

wherein:

S/R : the actual Signal-Interference-Ratio of each time slot ; SIR' : the estimated Signal-Interference-Ratio of each time slot; T : the accumulative time slot number. ξ : ξ = SIR - SIR' ;

As many apparently widely different embodiments of the present invention can be made without departing from the spirit and scope thereof, it is to be understood that the invention is not limited to the specific embodiments thereof except as defined in the appended claims.

Claims

1. A SINR estimation apparatus in a communications system, comprising: a demodulator for converting an I component and Q component of the received interfered signals into decided symbol; a modulator, coupled to the demodulator, for mapping the decided symbol into the QAM constellation point to obtain the estimated signal; delay means for delaying the I component and the Q component of the received interfered signal; subtraction means for subtracting the estimated signal from the received interfered signal to obtain the estimated interference; calculation means, coupled to the modulator and the subtraction means, for calculates the power of the estimated interference and the power of the estimated signal; and determination means for determining the SINR based on the power of the estimated interference the power of the estimated signals.

2. A SINR estimation apparatus in a communications system according to claim 1, further comprising accumulation average means for determining the accumulated values of the estimated signals and the estimated interference.

3. A SINR estimation apparatus in a communications system according to claim 1 or 2, further comprising a constellation power determination means, coupled to the SINR determination means, for determining the constellation power.

4. A SINR estimation apparatus in a communications system according to any one of claims 1 to 3 , wherein the SINR is determined by the following formula

wherein, SINR is SINR, S„, is the estimated signal power at the m-th symbol,

I_m the estimated interference power at the m-th symbol, T is the accumulation time

slots.

5. A SINR estimation method in a communications system, comprising the 5 steps of : converting an I component and a Q component of the received interfered signals into decided symbol; mapping the decided symbol into the QAM constellation point to obtain the estimated signal; 0 delaying the I component and the Q component of the received interfered signal; subtracting the estimated signal from the received interfered signal to obtain the estimated interference; calculating the power of the estimated interference and the power of the [5 estimated signal; and determining the SINR based on the power of the estimated interference the power of the estimated signals.

6. A SINR estimation method in a communications system according to claim 5, further comprising the step of determining the accumulated values of the 0 estimated signals and the estimated interference.

• 7. A SINR estimation method in a communications system according to claim 5 or 6, further comprising the step of determining the constellation power.

8. A SINR estimation method in a communications system according to any one of claims 5 to 7 , wherein the SINR is determined by the following formula

wherein, SINR is SIΝR, S_m is the estimated signal power at the m-th symbol, ϊ,„ the estimated interference power at the m-th symbol, T is the accumulation time slots.