KR20070020561A - SIR Estimation in Wireless Receiver - Google Patents

Abstract

The present invention relates to a method and apparatus for removing bias from an initiating signal to interference ratio (SIR). In an exemplary embodiment, the starting SIR calculator in the SIR processor calculates the starting SIR based on the signal received by the wireless receiver, while the average SIR calculator of the SIR processor generates an average SIR. Using the average SIR, the bias remover removes the bias from the starting SIR. In addition, according to an exemplary embodiment of the present invention, the bias canceller generates a scaling factor based on the average SIR and offset parameters, where the offset parameters are generated by the wireless receiver and a plurality of despread values processed by the wireless receiver. It is obtained from at least one of the plurality of paths of the plurality of path channels processed. In this embodiment, the bias remover includes a multiplier that increases the starting SIR by the scaling factor to remove the bias from the starting SIR.

Initiation signal-to-interference ratio, average signal-to-interference ratio, bias, offset parameter, scaling factor, multiplier

Description

SIR estimation in the wireless receiver {SIR ESTIMATION IN A WIRELESS RECEIVER}

The present invention relates generally to signal processing in a wireless network, and more particularly to estimating a signal to interference ratio (SIR) in a wireless receiver.

Receivers in a wireless network typically calculate performance parameters to evaluate the receiver and / or to assess certain network level parameters such as transmit power, data rate, and the like. One performance parameter of particular interest for spread spectrum networks is the signal to interference ratio (SIR) for the received signal. Conventional receivers typically calculate the SIR associated with the received signal and use the calculated SIR to adapt the network level parameters to the current channel conditions. For example, the calculated SIR may be used to control mobile station transmit power, data rate, mobile station scheduling, and the like.

The accuracy of network adaptation to current channel conditions depends not only on the accuracy of the SIR estimate but also on the amount of time used to generate the SIR estimate. In general, there are several ways to estimate SIR in spread spectrum networks. For example, the receiver can estimate the SIR using a combination of chip samples and despread symbols. While this approach provides accurate SIR estimates in an appropriate way, this approach requires complex receiver structures to access both chip samples and despread values.

Another receiver may estimate the SIR using the symbol estimates provided by the RAKE receiver. However, because the current RAKE output symbol corresponds to a signal that was well received in the past, the resulting SIR does not correspond to the current receive performance and channel conditions. Therefore, this approach requires a significantly less complex receiver structure, while the resulting SIR estimate is not sufficient for real-time operation such as power control, rate adaptation, and so on.

In addition, another receiver may use a despread symbol (pilot or data) to generate a finger SIR for each finger of the RAKE receiver. Summing the finger SIRs provides an SIR estimate that can be used for real time operation. However, since despread symbols typically contain a lot of noise to consider, deterministic SIR estimates are often biased. Conventional networks can eliminate this bias by subtracting the bias estimate from the current SIR estimate. However, the bias estimation process can overestimate bias. As a result, using subtraction to provide a bias has negative consequences and results in inaccurate SIR estimates.

The present invention is directed to a method and apparatus for removing bias from an onset estimate of a signal to interference ratio (SIR). In an exemplary embodiment, the SIR processor in the SIR estimator in the wireless receiver includes a starting SIR calculator, an average SIR calculator, and a bias canceller. The starting SIR calculator calculates the starting SIR based on the signal received by the wireless receiver, while the average SIR calculator generates an average SIR. Using the average SIR, the bias remover removes the bias from the starting SIR.

In an exemplary embodiment, the SIR estimator obtains the despread value from the baseband signal r (t). The SIR estimate uses the despread value to generate channel estimates and noise statistics, which in turn are used by the SIR processor to calculate the starting SIR estimate and the average SIR estimate.

In addition, according to an exemplary embodiment of the present invention, the bias remover generates a scaling factor based on the average SIR and offset parameters, where the offset parameters are obtained from a plurality of despread values processed by the wireless receiver. In this embodiment, the bias controller includes a counter that generates a scaling factor and a multiplier that increases the starting SIR by the scaling factor to remove the bias from the starting SIR.

1 illustrates an exemplary wireless network.

2 illustrates an exemplary wireless receiver of the wireless network of FIG.

3 is a block diagram of an SIR estimator.

4 is a block diagram of another SIR estimator.

5 is a block diagram of an exemplary SIR estimator in accordance with the present invention.

6 is a block diagram of an exemplary SIR processor for the SIR estimator of FIG.

6A and 6B are block diagrams of respective, exemplary signal and noise power estimators for the SIR processor of FIG.

6C is a block diagram of an example average SIR calculator for the SIR processor of FIG.

6D is a block diagram of an exemplary signal statistics estimator for the average SIR calculator of FIG. 6C.

6E is a block diagram of an exemplary bias controller for the SIR processor of FIG.

7 is a block diagram of another exemplary SIR processor for the SIR estimator of FIG.

FIG. 7A is a block diagram of an exemplary average SIR calculator for the SIR estimator of FIG.

8 is a block diagram of another exemplary average SIR processor for the SIR estimator of FIG.

8A is a block diagram of an example channel estimation processor for the SIR estimator of FIG.

FIG. 8B is a block diagram of an exemplary average SIR calculator for the SIR processor of FIG. 8. FIG.

8C is a block diagram of an exemplary bias canceller for the SIR processor of FIG.

9 is an exemplary flowchart of the invention.

1 illustrates an example spread spectrum wireless communication network 10. The wireless communication network includes one or more base stations 12, one or more mobile stations 20 and possibly one or more interfering objects 18. As used herein, the term "mobile station" refers to a cellular wireless telephone with or without a multi-line display; A personal communication system (PCS) terminal capable of combining data processing, facsimile and data communication capabilities with a cellular wireless telephone; Personal digital assistants (PDAs), which may include cordless phones, pagers, Internet / intranet access, web browsers, organizers, calendars, and / or global positioning system (GPS) receivers; And other appliances including conventional laptop and / or palmtop receivers or cordless telephone transmitters. The mobile station may also be called a "pervasive computing" device.

Base station 12 includes one or more antennas 14 that transmit a spread spectrum signal having one or more symbols to mobile station 20 and receive the signal from the mobile station. The transmitted signal typically includes traffic and pilot signals. Objects such as interfering object 18 cause multiple " eco " or delayed versions of the transmitted symbols to arrive at the mobile station 20 at different times. Receiver 16 processes a number of symbols at mobile station 20. Similarly, mobile station 20 may transmit symbols via one or more antennas 22 along various paths to base station 12, with receiver 16 processing multiple received symbol images.

)를 생성한 다. 심볼 추정치(

)는 필요로 된다면 부가적인 프로세서(28)에서 더 처리된다. 예를 들어, 부가적인 프로세서(28)는 터보 디코더(도시되지 않음)를 포함할 수 있는데, 이는 기저대 프로세서(30)에 의해 제공된 심볼 추정치에 기초하여 정보 비트 값을 결정한다. 이런 정보 비트 값은 음성, 이미지 등으로 변환될 수 있다.2 shows an example receiver 16 for a base station and / or mobile station 20. Receiver 16 receives and processes the symbols of the received signal to generate a received symbol estimate. Exemplary receiver 16 includes a receiver front end 26, baseband processor 30, and additional processor 28. Receiver front end 26 typically includes a conversion circuit such as a filter, mixer and / or analog-to-digital converter to generate a series of digitized baseband signal samples r (t) corresponding to the received signal. . The baseband processor 30 demodulates the baseband signal r (t) to obtain a symbol estimate corresponding to the received signal (

Create). Symbol estimates (

Is further processed in additional processor 28, if necessary. For example, additional processor 28 may include a turbo decoder (not shown), which determines information bit values based on symbol estimates provided by baseband processor 30. This information bit value can be converted to voice, image, or the like.

)의 벡터를 발생시킨다. 게다가, 잡음 파워 추정기(40)는 역확산 심볼(y) 및 채널 추정치(c)에 기초하여 잡음 파워 추정치(

)의 벡터를 발생시킨다. 분리기(42)는 잡음 파워 추정치(

)의 벡터에서의 상응하는 성분에 의해 신호 파워 추정치(

)의 벡터에서의 각각의 성분을 분리하여 개시 핑거 SIR 추정치의 벡터(SIR_init)를 발생시킨다. 개시 핑거 SIR 추정치(SIR_init)의 벡터 성분들은 누산기(44)에 누산되어 기저대 신호(r(t))에 상응 하는 개시 SIR 추정치(SIR_final)를 생성한다. 도3의 SIR 추정기(32a)는 실시간 동작 동안에 적합한 SIR_final를 발생시키는 반면, SIR_final을 발생시키기 위해 사용되는 잡음 채널 추정치에 의해 야기된 바이어스를 고려하지 않는다.As shown in FIG. 2, the baseband processor 30 may include a signal-to-interference ratio (SIR) estimator 32 that estimates the SIR from the baseband signal r (t). 3 shows an SIR estimator 32a, which despreads a unit 34, a channel estimator 36, a magnitude squared calculator 38, a noise power estimator 40, a separator 42, and an accumulator 44. Include. Despreading unit 34 despreads the received signal to generate a vector of despread symbols and values y. Each vector component of despread symbol y corresponds to various timing offsets for different signal paths of multiple channels. Based on each vector component of despread symbol y, channel estimator 36 generates a vector of channel estimates c. The magnitude squared calculator 38 calculates signal power estimates based on the respective vector components of the channel estimate c.

Generates a vector of In addition, the noise power estimator 40 is based on the despread symbol y and the channel estimate c.

Generates a vector of Separator 42 is a noise power estimate (

Signal power estimate (i) by the corresponding component in the vector of

Each component in the vector of N +) is separated to generate a vector SIR _init of the starting finger SIR estimate. The vector components of the starting finger SIR estimate SIR _init are accumulated in the accumulator 44 to produce an starting SIR estimate SIR _final corresponding to the baseband signal r (t). SIR estimator 32a of FIG. 3 generates a suitable SIR _final during real-time operation, while generating the SIR _final . Used to generate It does not take into account the bias caused by the noise channel estimate.

One way to reduce the bias associated with the SIR estimate is to reduce the noise present in the channel estimate (c) used to generate the SIR estimate. At low Doppler spreads this can be achieved by smoothing the channel estimate c over time. However, time sensitive network operation with accurate SIR estimates often cannot wait for the amount of time needed to smooth the channel estimate (c). As such, such a method is not useful for time sensitive operation.

) 및 잡음 파워 추정치(

)를 각각 발생시킨다. 그러나 도3의 SIR 추정기와는 다르게, 도4의 SIR 추정기(32a)는 신호 결합기(48)의 신호 파워 추정치(

)로부터 신호 바이어스 추정 치를 감산함으로써 신호 바이어스를 제거하여 수정된 신호 파워 추정치(

)를 발생시킨다. 유사하게는, 잡음 결합기(52)가 잡음 파워 추정치(

)로부터 잡음 바이어스의 추정치를 감산하여 수정된 잡음 파워 추정치(

)를 발생시킨다. 분리기(54)는 수정된 잡음 파워 추정치(

)에 의해서 수정된 신호 파워 추정치(

)를 분리하여 기저대 신호(r(t))에 상응하는 최종 SIR 추정치(SIR_final)를 발생시킨다. 이런 실시예가 바이어스를 고려하지만, 이런 실시예는 또한 신호 또는 잡음 바이어스 텀(terms)을 과대 평가함으로써 야기되는 오류를 범하는데, 이는 부정적인 SIR_final의 결과를 가져올 수 있다. Another way to reduce the bias is to generate a starting SIR estimate and remove the bias from the starting SIR estimate. WO 01/65717, entitled “Correction of Received Signal and Interference Estimates,” incorporated herein by reference, discloses an SIR estimator 32 that can be used to remove bias from the starting SIR estimate. 4 shows such an SIR estimator 32a. In this embodiment, the SIR estimator 32a includes a signal power estimator 46, a signal combiner 48, a noise power estimator 50, a noise combiner 52, and a separator 54. The signal power estimator 50 and the noise power estimator 56 directly estimate the signal power estimate () from the baseband signal r (t).

) And noise power estimate (

) Respectively. However, unlike the SIR estimator of FIG. 3, the SIR estimator 32a of FIG. 4 uses the signal power estimate of the signal combiner 48.

The signal power estimate modified by subtracting the signal bias estimate from

). Similarly, noise combiner 52 provides noise power estimate (

Subtracting the estimate of the noise bias from

). Separator 54 is a modified noise power estimate (

Signal power estimate modified by

) Is generated to generate a final SIR estimate (SIR _final ) corresponding to the baseband signal r (t). Although this embodiment takes into account bias, this embodiment also commits an error caused by overestimating the signal or noise bias terms, which can result in a negative SIR _final .

5 shows a block diagram of an exemplary SIR estimator 32b in accordance with the present invention. SIR estimator 32b provides a final SIR estimate (SIR _fanal ), which takes into account bias without introducing subtraction errors caused by previous SIR estimation approaches. SIR estimator 32b includes a despread unit 34, a channel estimator 36, a noise statistics estimator 56, and an SIR processor 100. The despreading unit 34 despreads the baseband signal r (t) to generate a vector of despread symbols y. As those skilled in the art will appreciate, each component of the despread symbol vector corresponds to various timing offsets associated with the various signal paths of the multi-path channel. Based on the despread symbol, channel estimator 36 generates a vector of channel estimates c according to various means known in the art. For example, the channel estimate vector c is:

Where K denotes the number of pilot signals processed by receiver 16, b (i) denotes a known pilot symbol for the i ^th symbol period, and b * (i) denotes b ( i) represents a complex change in i) and y (i) represents a vector of despread symbols or values from various path delays during the i ^th symbol period.

The despreading value y along with the channel estimate c is also provided to the noise statistics estimator 56. Noise statistics estimator 56 estimates noise statistics between despread symbols y from various path delays. The noise statistic may be any statistic that represents the noise component of spread symbols y, such as quadratic statistics or correlation between noise on the despread symbol. As one of ordinary skill in the art recognizes that "covariance" is a specific case of "cross-correlation" with a meaning of zero, vocabulary such as "correlation" and "covariance" as used herein may be used to refer to a particular phrase. It should be recognized that context can be used interchangeably unless there is a clear distinction between the two vocabularies.

In an exemplary embodiment, the noise statistics estimator 56 estimates the correlation matrix M between the impairments on the despread symbols according to either of equations 2A or 2B:

Where the superscript "H" denotes a conjugate transpose. The speech statistic matrix, referred to herein as the noise correlation matrix R _x , can be obtained, for example, by setting the same R _N to M. Alternatively, the noise correlation matrix R _N may be obtained by smoothing the previous M value and using the exponential filter to set the same R _N to the smoothed M. Since M and R _N are Hermitian symmetric, it will be appreciated that only the top or bottom three of these matrices have to be computed, which greatly simplifies computational complexity.

Those skilled in the art will appreciate that the present invention is not limited to the noise statistics calculation method described above. In practice, the noise correlation matrix R _N can be calculated according to any means of the prior art. Exemplary methods are described in US Patent Specification Serial No. 10/811699, filed March 29, 2004, entitled "Method and Apparatus for," filed March 12, 2004. Parameter Estimation in a Generalized RAKE Receiver, disclosed in US Pat. No. 10/800167, both of which are incorporated herein by reference.

SIR processor 100 obtains SIR _final from channel estimate c and noise correlation matrix R _N as described below. Due to the time sensitive nature of the SIR estimate during some network operation, it will be appreciated that the channel estimate c relates to a value that can be formed using short term data for the purpose of the SIR estimation. Therefore, the SIR estimate channel estimate c may be different from the channel estimate calculated for the demodulator, where time delay is not important. As a result, a demodulator (not shown) of baseband processor 30 may use a multiple channel estimator that generates multiple channel estimates, for example, based on long term data. Although the present invention shows an SIR processor 100 using multiple channel estimates in addition to those used in a demodulator, those skilled in the art will simplify the receiver architecture by sharing the channel estimates provided by the SIR processor 100 and the demodulator with a single channel estimator. It will be recognized.

6 illustrates an exemplary embodiment of an SIR processor 100 in accordance with the present invention. The SIR processor of FIG. 6 assumes that noise on despread symbol y is related to each other, which occurs due to filter or distributed channel interference at receiver front end 26. SIR processor 100 includes an initiating SIR calculator 102, an average SIR calculator 104, and a bias remover 106. The starting SIR calculator 102 calculates the weight calculator 108, the signal power estimator 110, and the noise power estimator 116 to obtain an starting SIR estimate SIRinit based on the channel estimate c and the noise correlation matrix R _N. ) And a separator 120.

As such, the weight calculator 108 calculates a vector of weighting factors w based on the channel estimate c in accordance with any known method. For example, when receiver 16 includes a conventional RAKE receiver, the weighting factor w is approximated according to equation (3):

.

.

However, when receiver 16 includes a generalized RAKE (G-RAKE) receiver, weight calculator 108 uses both the channel estimate (c) and noise correlation matrix (R _N ) to weight the factors according to the following equation: (w) can be calculated:

.

.

(Interested readers are entitled "A Generalized RAKE Receiver for Interference Suppression" to G. Bottomley, T. Ottosson and Y.-PEWang, published in IEEE 2000 Selected Areas Communications, 18: 1536-1545, August 2000. Alternatively, the weighting factor w may be referred to as, for example, “Method and Apparatus for RAKE Receiver Combining Weight Generation”, filed Sep. 26, 2003. Can be calculated by other methods disclosed in US Pat. No. 10/672127, which is incorporated herein by reference. According to this method, the weighting factor w can be calculated by the following equation:

Where F depends on channel and noise statistics. In some cases, along with the channel estimates, it will be appreciated that the weighting factor w relates to values that can be formed using short term data for the purpose of SIR estimation. Therefore, the SIR estimated weighting factor w may be different from the weighting factor calculated for the demodulator, where the time delay is not important. As a result, the SIR processor 100 of the present invention includes a weight calculator 108 that can reference a weighting factor w other than that used by the demodulator. However, those skilled in the art will appreciate that the SIR estimator 32b and demodulator can share the weighting factor provided by a single weight calculator to simplify the receiver structure.

)를 발생시킨다. 도6A는 예시적인 신호 파워 추정기(110)를 도시한다. 신호 파워 추정기(110)는 내부 프로덕트 계산기(112) 및 크기 제곱 계산기(114)를 포함한다. 크기 제곱 계산기(114)는 내부 프로덕트 계산기(112)에 의해 제공된 웨이팅 팩터(w) 및 채널 추정치(c)의 내부 프로덕트의 크기를 제곱하여 수학식6에서 보여지는 바와 같이 신호 파워 추정치(

)를 발생시킨다:Based on the calculated weighting factor w, the starting SIR calculator 102 calculates the signal and noise power estimates used to compute the SIR _init . In particular, the signal power estimator 110 is based on the channel estimate c and the weighting factor w, according to any means known in the art.

). 6A shows an example signal power estimator 110. Signal power estimator 110 includes an internal product calculator 112 and a magnitude squared calculator 114. The magnitude squared calculator 114 squares the magnitudes of the weighting factor w provided by the internal product calculator 112 and the internal product of the channel estimate c to obtain the signal power estimate (

Generates):

.

.

)를 발생시킨다. 예시적인 잡음 파워 추정기(116)는 도6B에 도시된 바와 같이 2차 컴퓨터(118)를 포함하는데, 이는 다음의 수학식 7에 따라 잡음 상관 행렬(R_N) 및 웨이팅 팩터(w)로부터 잡음 파워 추정치(

)를 얻을 수 있다:The noise power estimator 116 turns on the overall noise power estimate based on the noise correlation matrix R _N and the weighting factor w according to any means known in the art.

). Exemplary noise power estimator 116 includes a secondary computer 118 as shown in FIG. 6B, which is noise power from noise correlation matrix R _N and weighting factor w according to Equation 7 below. Estimate (

You can get

.

.

)에 의해 신호 파워 추정치(

)를 분리하여 개시 SIR 추정치(SIR_init)를 발생시킨다. 바이어스 제거기(106)는 평균 SIR 계산기(104)에 의해 발생된 평균 SIR 추정치(

)를 사용하여 SIR_init으로부터 바이어스를 제거함으로써 SIR_init를 또한 리파인(refine)한다. 바람직한 실시예에서, 바이어스 제거기(106)는 평균 SIR 추정치(

)로부터 얻어진 스케일링 팩터(

)에 의해 개시 SIR 추정치(SIR_init)를 증가시키는 배율기를 포함한다. 평균 SIR에 기초한 스케일링 팩터(

)에 의해 SIR_init을 증가시킴으로써, 바이어스 제거기(106)는 개시 SIR 추정치(SIR_init)로부터 바이어스를 제거하여 최종 SIR 추정치(SIR_final)를 발생시킨다. Separator 120 has a noise power estimate (

Signal power estimate (

) Is generated to generate a starting SIR estimate (SIR _init ). The bias remover 106 is an average SIR estimate generated by the average SIR calculator 104.

) Also refined (refine) the SIR _init by removing bias from a SIR _init using. In a preferred embodiment, the bias remover 106 is a mean SIR estimate (

Scaling factor obtained from

A multiplier for increasing the starting SIR estimate (SIR _init ) by. Scaling factor based on average SIR (

By increasing the SIR _init , the bias remover 106 removes the bias from the starting SIR estimate (SIR _init ) to generate a final SIR estimate (SIR _final ).

6C shows an example average SIR calculator 104 for the SIR processor 100 of FIG. The average SIR calculator 104 includes a signal statistics estimator 122, a signal secondary computer 124, a noise secondary computer 126, and a separator 128. 6C shows separate secondary computers 124, 126 that calculate the average and noise signals, respectively, although those skilled in the art can combine these secondary computers 124, 126 into a single secondary computer that calculates both the average signal and the noise power. It will be recognized.

The signal statistics estimator 122 calculates a signal correlation matrix Q based on the channel estimate c and the noise correlation matrix R _N. An exemplary signal statistics estimator 122 is shown in FIG. 6D. Signal statistics estimator 122 includes an external product calculator 132, a smoothing filter 134, a multiplier 136, and a combiner 138. Smoothing filter 134 smoothes the external product of channel estimate c, provided by external product calculator 132 over time, to generate a channel estimation correlation matrix P as shown in equation (8):

,

,

Where E {} represents the expected value. Since P is the mission symmetry, it will be appreciated that only the top or bottom three of the channel estimation correlation matrix must be computed, which greatly simplifies the computational complexity of the present invention.

Since the channel estimate c includes noise due to estimation error, the channel estimation correlation matrix P represents the biased estimate of the signal correlation matrix Q. Therefore, to remove the bias, the signal statistics estimator 122 subtracts, at the combiner 138, the scaled version of the noise correlation matrix provided by the multiplier 136 to obtain a signal correlation matrix as shown in equation (9). Generates Q):

,

,

Where β depends on the number K of despread symbols used to estimate the vector of channel estimate c. K may also include the associated power or energy level between pilot and traffic data. When K is large, or when there is a benefit to simplify the operation, Q is set equal to P.

)를 계산한다:The signal statistics estimator 122 provides the signal correlation matrix Q to the signal secondary computer 124, which is an average signal power according to equation (10).

Calculate)

.

.

)를 계산한다:Similarly, the noise secondary computer 126 uses the noise correlation matrix (RN) to obtain the average noise power (Eq.

Calculate)

.

.

)에 의해서 평균 신호 파워(

)를 분리함으로써 평균 SIR을 발생시킨다.Separator 128 has an average noise power (

), Average signal power (

) To generate an average SIR.

)로부터 스케일링 팩터(

)를 얻는다:FIG. 6E illustrates an example bias remover 106 for the SIR processor 100 of FIG. 6. The bias remover 106 includes a converter 130 and a multiplier 131. The converter 130 calculates an average SIR estimate provided by the output of the average SIR calculator 104 according to the following equation.

From the scaling factor (

To get:

,

,

)를 사용하여 최종 SIR 추정치(SIR_final)를 발생시킨다. Where α represents an offset parameter obtained from the number K of despread symbols used to generate the channel estimate c. K may also include the associated power or energy level between pilot and traffic data. In a preferred embodiment, the offset parameter α can be calculated with the equation α = 1 / K. Multiplier 131 removes the bias from the starting SIR estimate (SIR _init ) and provides a scaling factor provided by converter 130 (

) To generate a final SIR estimate (SIR _final ).

Returning to FIG. 7 again, another embodiment of an example SIR processor 100 will be disclosed. Like the SIR processor 100 of FIG. 6, the SIR processor 100 of FIG. 7 includes an initiating SIR calculator 150, a bias remover 154, and an average SIR calculator 156. In this example, it is assumed that a G-RAKE receiver is used, where the weighting factor is calculated according to equation (4). Thus, the initiating SIR calculator 150 can use the inverse secondary computer 152 to calculate the initiating SIR estimate SIR _init according to the following equation:

.

.

를 획득하는데 사용될 수 있다.Note: Those skilled in the art recognize that there are several ways to simplify the calculation. For example, Gauss-Seidel first

Can be used to obtain.

)를 사용하여 SIR_init로부터 바이어스를 제거한다. 도7A는 도7에 도시된 SIR 프로세서(100)를 위한 예시적인 평균 SIR 계산기(156)를 도시한다. 평균 SIR 계산기(156)는 신호 통계 추정기(158)를 포함하는데, 이는 상술된 바와 같이 신호 상관 행렬(Q)을 발생시킨다. 평균 SIR 계산기(156)는 평균 SIR 프로세서(160)를 또한 포함하는데, 이는 신호 상관 행렬(Q) 및 잡음 상관 행렬(R_N)을 사용하여 다음 수학식에 따라 평균 SIR 추정치(

)를 계산한다:In conjunction with the first embodiment, the bias remover 154 provides an average SIR estimate (provided by the average SIR calculator 156).

) To remove the bias from the SIR _init . FIG. 7A shows an exemplary average SIR calculator 156 for the SIR processor 100 shown in FIG. The average SIR calculator 156 includes a signal statistics estimator 158, which generates a signal correlation matrix Q as described above. The average SIR calculator 156 also includes an average SIR processor 160, which uses a signal correlation matrix (Q) and a noise correlation matrix (R _N ) to calculate the average SIR estimate according to the following equation:

Calculate)

,

,

Where Tr {} represents the trace of the product of R _N −1 and Q. Traces in a matrix can be computed in any number of ways. For example, one method of computing a trace of the product of R _N -1 and Q calculates the product of the columns by calculating the following equation:

,

,

Where q is the column of Q. Equation 15 may be solved using, for example, a Gauss-Sidel or Gauss-Jordan interaction approach.

)는 지연 경로 SIR 값의 합에 가까워질 수 있는데, 여기서 지연 경로 SIR 값은 수학식 16에서 보여지는 바와 같이, R_N(지연 경로 잡음 파워)의 대각 성분에 의해서 Q(지연 경로 신호 파워)의 대각 성분을 분리함으로써 획득된다:Another method for solving Equation 15 and approximating the trace of the product of R _N ^-1 and Q assumes that the noise correlation matrix R _N and the signal correlation matrix Q are diagonal matrices. This can be accomplished by ignoring the non-diagonal matrices of R _N and Q and / or setting the non-diagonal components to zero. In both cases, the average SIR estimate (

) Can be approximated to the sum of the delay path SIR values, where the delay path SIR value is determined by the diagonal component of R _N (delay path noise power), as shown in Eq. Obtained by separating diagonal components:

,

,

)에 대한 이런 접근법은 기저대 프로세서가 통상적인 RAKE 수신기를 포함하고 여러 역확산 심볼에 존재하는 잡음이 정확하지 않을 때 특히 효과적이다. 잡음 상관 행렬(R_N)이 대각 행렬이고 또한 SIR_init 계산을 간단하게 한다는 가정이 인식될 것이다. 결과적으로, 수학식 13은 지연 경로 SIR 값의 합에 가까워질 수 있고, 여기서 각각의 핑거는 자신의 평균 신호 및 잡음 파워를 갖는다.Where J represents the number of delay paths processed by the receiver. Since the non-diagonal components are ignored (or set to zero), the non-diagonal components of Q need not be calculated, which saves processing time. Estimate (

This approach is particularly effective when the baseband processor contains a typical RAKE receiver and the noise present in the various despread symbols is not accurate. It will be appreciated that the noise correlation matrix R _N is a diagonal matrix and also simplifies the SIR _init calculation. As a result, Equation 13 can be close to the sum of the delay path SIR values, where each finger has its own average signal and noise power.

은 다음 수학식에 따라 컴퓨트될 수 있다:In addition, another method of approaching the trace of the product of R _N ^-1 and Q is that the noise correlation matrix R _N is a diagonal matrix, and the noise associated with each delay path processed by the receiver is statistical noise, so that the same noise power It is assumed to have (N). As a result, the diagonal components of the noise correlation matrix R _N are the same. therefore,

Can be computed according to the following equation:

,

,

Where J represents the number of delay paths or fingers processed by the receiver. It will be appreciated that the noise correlation matrix R _N is a diagonal matrix and also simplifies the SIR _init calculation. As a result, equation (13) simplifies the sum of the finger signal power values separated by the noise power (N).

가 계산되면, 바이어스 제거기(154)는 평균 SIR 추정치(

)FMF 사용하여 개시 SIR 추정치(SIR_init)로부터 바이어스를 제거한다. 예시적인 실시예에서, 바이어스 제거기(154)는

로부터 비롯된 스케일링 팩터(

)에 의해 SIR_init을 증가시키는 배율기를 포함한다. 이런 예에서, 바이어스 제거기(154)는 다음 수학식 에 따라, 도6E의 컨버터(130)에서

에 기초한 스케일링 팩터(

)를 컴퓨트할 수 있다:First

Is calculated, the bias remover 154 calculates an average SIR estimate (

Remove the bias from the starting SIR estimate (SIR _init ) using FMF. In the exemplary embodiment, the bias remover 154 is

Scaling factor from

) And a multiplier to increase the SIR _init . In this example, the bias remover 154 is in the converter 130 of Fig. 6E, according to the following equation.

Scaling factor based on

) Can be computed:

,

,

Where α is an offset parameter that can be calculated as α = J / K, J represents the number of delay paths processed by the receiver, and K represents the number of symbols used to calculate the channel estimate c.

8 illustrates an example embodiment of another SIR processor 100. This embodiment uses a schematic form of joint scaling, which is an extension of G-RAKE used to account for noise channel estimates when forming and combining weights. As in the previous embodiment, the SIR processor 100 of FIG. 8 also includes a starting SIR calculator 170, an average SIR calculator 180, and a bias remover 190. However, in this embodiment, the starting SIR calculator 170 refines the channel estimate c before calculating the SIR _init .

As a result, the starting SIR calculator 170 includes a channel estimation processor 174 and an inverse secondary computer 172. Channel estimation processor 174 as shown in FIG. 8A includes a channel estimation matrix calculator 176 and a matrix multiplier 178. The channel estimation matrix calculator 176 computes the channel estimation matrix A according to the following equation:

,

,

Where K represents the number of despread symbols used to calculate the channel estimate (c) and may also include the effect of the power level difference between pilot and traffic data. Since the channel estimation matrix A depends on the signal statistics Q and the noise statistics R _N , as shown in equation 19, the channel estimation matrix A has a minimum mean square error (MMSE). ) Provides a form of channel estimation.

)를 발생시킨다. 개시 SIR 계산기(170)는 그 후에 아래의 수학식 20에서 보여지는 바와 같이, 수정된 채널 추정치(

)를 사용하여 역 2차 컴퓨터(172)에서 개시 SIR 추정치(SIR_init)를 계산한다:The channel estimation processor 174 refines the original channel estimate c by applying the channel estimation matrix A to the channel estimate c of the matrix multiplier 178, thereby modifying the original channel estimate c.

). The starting SIR calculator 170 then revises the modified channel estimate (< / RTI >) as shown in Equation 20 below.

Calculate the starting SIR estimate (SIR _init ) at inverse secondary computer 172 using:

.

.

)를 생성하는 채널 추정 행렬(A)에 의한 채널 추정치(c)의 수정이다. As shown in Equation 20, the embodiment of FIG. 8 is similar to the example of FIG. The primary difference is the modified channel estimate (

Is a correction of the channel estimate (c) by the channel estimation matrix (A) that generates.

)를 수정한다. 도8B는 도8의 SIR 프로세서(100)를 위한 예시적인 평균 SIR 계산기(180)를 도시한다. 도7A의 평균 SIR 계산기(156)와 같이, 도8B의 평균 SIR 계산기(180)는 신호 통계 추정기(182)를 포함한다. 게다가, 평균 SIR 계산기(180)는 행렬 제곱 계산기(184) 및 도7A의 평균 SIR 프로세서(160)를 대신하는 수정된 평균 SIR 프로세서(186)를 포함한다. 도8B의 실시예에서, 수정된 평균 SIR 프로세서(186)는 두 개의 평균 SIR 추정치(

,

)를 컴퓨트한다. 제1 평균 SIR 추정치(

)는 수학식 14으로 계산되는데, 이는 여기서 수학식 21과 같이 반복된다:In addition, the channel estimation matrix A also provides an average SIR estimate (

). FIG. 8B shows an exemplary average SIR calculator 180 for the SIR processor 100 of FIG. 8. Like the average SIR calculator 156 of FIG. 7A, the average SIR calculator 180 of FIG. 8B includes a signal statistics estimator 182. As shown in FIG. In addition, the average SIR calculator 180 includes a matrix square calculator 184 and a modified average SIR processor 186 replacing the average SIR processor 160 of FIG. 7A. In the embodiment of FIG. 8B, the modified average SIR processor 186 may include two average SIR estimates (

,

Compute). First mean SIR estimate (

) Is calculated by Equation 14, where it is repeated as Equation 21:

.

.

)는 행렬 제곱 계산기(184)에 의해서 제공되는 채널 추정 행렬(A)뿐만 아니라 신호 상관 행렬(Q) 및 잡음 상관 행렬(R_N)의 제곱에 의해서, 수학식 22에서 보여지는 바와 같이 얻어진다:Second average SIR estimate (

Is obtained by the square of the signal correlation matrix (Q) and the noise correlation matrix (R _N ) as well as the channel estimation matrix (A) provided by the matrix square calculator (184), as shown in equation (22):

.

.

는 채널 추정 행렬(A)에 따르는 스케일링 다운 팩터(scaling down factor)를 필수적으로 갖는데, 이는 직관적으로 채널 추정치(c)에서의 잡음을 고려한다.As shown in Equation 22,

Has essentially a scaling down factor according to the channel estimation matrix A, which intuitively takes into account the noise in the channel estimate c.

및

을 사용하여 SIR_init로부터 바이어스를 제거한다. 바 이어스 제거기(190)는 컨버터(188) 및 배율기(189)를 포함한다. 평균 SIR 계산기(180)에 의해 제공되는

및

을 사용하면, 컨버터(188)는 다음 공식에 따라 스케일링 팩터(

)를 컴퓨트 할 수 있다:FIG. 8C shows an example bias remover 190 for the SIR processor 100 DMF of FIG. 8, which

And

Use to remove the bias from the SIR _init . The bias remover 190 includes a converter 188 and a multiplier 189. Provided by the average SIR calculator 180

And

Using the converter 188, the converter 188 can calculate the scaling factor according to

Can compute):

,

,

로 계산될 수 있다. 일부 실시예에서,

및

에 관련된 컴퓨테이션(computation)을 간단하게 하기를 희망할 수 있다. 그 때문에, A^HA는 단위 행렬로서 근사치가 구해지고,

및 는 도7을 참조하여 상기 설명된 방법에 따라 컴퓨트될 수 있다. 어떤 경우에는, 배율기(189)가 SIR_init로부터 바이어스를 제거하여 컨버터(188)에 의해 제공된 스케일링 팩터(

)를 사용하여 개시 SIR 추정치(SIR_init)를 스케일링함으로써 최종 SIR 추정치(SIR_final)를 발생시킨다. Where α is

It can be calculated as In some embodiments,

And

You may wish to simplify the computation associated with. Therefore, A ^H A is approximated as the unit matrix,

And Can be computed according to the method described above with reference to FIG. In some cases, the multiplier 189 removes the bias from the SIR _init to provide a scaling factor provided by the converter 188.

) To generate a final SIR estimate (SIR _final ) by scaling the starting SIR estimate (SIR _init ).

)를 발생시킨다(블록210). 평균 SIR 추정 치(

)를 사용하면, SIR 프로세서(100)는 개시 SIR 추정치(SIR_init)로부터 바이어스를 제거하여 최종 SIR 추정치(SIR_final)를 발생시킨다(블록220).A method and apparatus for calculating a final SIR estimate (SIR _final ) by removing bias from the starting SIR estimate (SIR _init ) above. 9 shows an exemplary method of implementing the present invention. In accordance with the present invention, the SIR processor 100 calculates a starting SIR estimate based on the received signal (block 200). In addition, the SIR processor 100 may determine an average SIR estimate based on the received signal.

(Block 210). Average SIR Estimates (

), The SIR processor 100 removes the bias from the starting SIR estimate SIR _init to generate a final SIR estimate SIR _final (block 220).

Although the SIR processor 100 of the present invention is shown having several separate components, those skilled in the art recognize that two or more such components may be combined in the same functional circuit. In addition, those skilled in the art will recognize that one or more such circuits may be implemented in hardware and / or software (including firmware, software, micro-code, etc.) to include custom semiconductors, field programmable gate arrays, and the like. Software or code for implementing the present invention may be stored on any known computer readable medium.

)는 평균 SIR 추정치(

)로부터 얻어지고, 차례로 기저대 신호(r(t))로부터 얻어진다. 결과적으로, 본 발명의 SIR 프로세서(100)는 평균 SIR 추정치(

)를 사용하여 SIR_init에 존재하는 바이어스를 제거한다.

를 계산하는 몇몇 방법이 상술되었지만, 본 발명은 이런 방법에 국한되지 않는다. 예를 들어,

은 이전 최종 SIR 값을 평활화함으로써 근사치가 구해질 수 있다. 게다가, 파워 제어 루프, 명목상의 SIR 또는 최악의 경우 SIR에 사용되는 타겟 SIR이

로 한정될 수 있고, 스케일링 팩터(

)를 계산하는데 사용될 수 있다. As shown in Figs. 6-8, in the above description, the scaling factor (

) Is the average SIR estimate (

Is obtained from the baseband signal r (t). As a result, the SIR processor 100 of the present invention provides an average SIR estimate (

) To remove the bias present in the SIR _init .

Although some methods for calculating the above have been described above, the present invention is not limited to this method. E.g,

Can be approximated by smoothing the previous final SIR value. In addition, the target SIR used for the power control loop, nominal SIR, or worst case SIR,

May be defined as a scaling factor (

) Can be used to calculate

)로 개시 SIR 추정치(SIR_init)를 증가시킴으로써 부정적인 최종 SIR 추정치(SIR_final)의 문제점을 피할 수 있다. 사전 테스트는 통상적인 감산 접근법에 비해 본 발명의 곱셈 접근법이 표준 RAKE 수신기에 대해 20% 정도 최종 SIR 추정치의 정확성을 증가시킬 수 있다고 보여진다. 2003년 9월 2일자로 출원된 "Method and Apparatus for Finger Placement in a DS-CDMA RAKE Receiver"라는 명칭의 미합중국 특허 명세서 일련번호 제 10/653679호에 개시된 바와 같이 핑거 교환을 위해서 그리드 접근법을 사용하는 수신기에서 정확성이 더 많이 개선되었다(40%-70%). 결과적으로, 본 발명은 시간 민감성 동작을 위해 정확한 최종 SIR 추정치를 제공하는 개선된 방법 및 장치를 개시한다. Since the methods and apparatus described above can immediately and individually compute the SIR _final , the resulting final SIR estimate (SIR _final ) can be used for real-time operation such as power control, rate adaptation, and the like. In addition, unlike the previous solution, the present invention provides a scaling factor for removing bias.

By increasing the starting SIR estimate (SIR _init ), the problem of negative final SIR estimate (SIR _final ) can be avoided. Pretesting has shown that the multiplication approach of the present invention can increase the accuracy of the final SIR estimate by 20% for a standard RAKE receiver compared to a conventional subtraction approach. Using a grid approach for finger exchange as disclosed in US Patent Specification Serial No. 10/653679, entitled "Method and Apparatus for Finger Placement in a DS-CDMA RAKE Receiver," filed September 2, 2003. More accuracy is improved at the receiver (40% -70%). As a result, the present invention discloses an improved method and apparatus for providing accurate final SIR estimates for time sensitive operation.

)를 계산하기 위해 사용되는 평균 SIR 추정치로서 장기간 SIR 추정치를 사용할 수 있다. While calculating the SIR _final for real time operation above is described, those skilled in the art will appreciate that the final SIR estimate may also be used to determine long term SIR. For example, the final SIR estimate computed over one or more frames can be averaged to generate a long term SIR estimate. This long term SIR estimate may be provided to the base station and other network entities as a long term quality measure. In addition, once the long-term SIR estimate is computed, the SIR processor 100 can determine the scaling factor (

The long term SIR estimate can be used as the average SIR estimate used to calculate.

The above also discloses the present invention with respect to the despread symbol y obtained from the baseband signal r (t). Those skilled in the art will appreciate that such despread symbols may be based on pilot channels that are treated as a series of pilot symbols, data symbols, and / or pilot symbols. In addition, the number of symbols and / or the type of symbols may be selectively changed based on the current channel state. For example, the Doppler spread estimator can be used to determine how to quickly switch channels. If the channel changes quickly, then a slot can be used, for example, from only a single time slot. If the channel changes slowly, symbols from multiple past slots can be used. For slowly changing channels, channel estimator 36 and / or noise stat estimator 56 can exponentially weight the contribution from the older slots and complement each slot for changes in transmit power due to power control. have.

Although the wireless network is described with respect to a single transmit and / or receive antenna, the present invention is not limited thereto and may be adapted to a network having multiple transmit and / or receive antennas. In this case, the fingers of the despread spectrum receiver are assigned to any path from any antenna. Therefore, vector quantities such as despread symbols, channel estimates, etc. can also be collected in the vector. However, for many transmit / receive antenna systems, the vector component has both a path and an antenna index. For example, when there are two receive antennas where each antenna receives signals in two different paths, the vector quantity consists of four lengths and is a 4x4 matrix. In addition, for many transmit antenna systems where several scrambled despread codes are used on several transmit antennas, it is suitably suitable to assume that the noise and fading terms are not accurate. As a result, the SIR processor 100 may generate and generate a starting SIR estimate for each transmitted signal separately and then sum the individual starting SIR estimates to obtain a full starting SIR estimate (SIR _init ). In this scenario, bias removal may occur before or after summation. It will also be appreciated that the present invention can be used with a transmit diversity system.

Of course, the invention may be carried out in a manner different from that specifically described herein without departing from the main features of the invention. Embodiments of the present invention may be considered in all respects as shown, but are not limited to, and it is intended that all changes within the scope and meaning of the appended claims may be included therein.

Claims (46)

A method of removing bias from an initiation signal to interference ratio generated by a wireless receiver, the method comprising: Calculating the initiation signal-to-interference ratio based on the signal received by the wireless receiver; Generating an average signal to interference ratio; And And removing the bias from the initiation signal to interference ratio using the average signal to interference ratio. 2. The wireless receiver of claim 1, wherein generating the average signal to interference ratio includes calculating the average signal to interference ratio based on channel estimates and noise statistics obtained from the received signal. A method for bias removal from the generated starting signal to interference ratio. The method of claim 2, Calculating the average signal-to-interference ratio based on channel estimates and noise statistics obtained from the received signal to calculate the average signal-to-interference ratio based on channel estimates and noise statistics obtained from despread values of the received signal. And biasing the starting signal-to-interference ratio generated by the wireless receiver. The method of claim 2, Using the average signal to interference ratio to remove the bias from the starting signal to interference ratio includes: Generating a scaling factor based on the average signal to interference ratio; And Increasing the initiation signal-to-interference ratio from the scaling factor to remove the bias. The method of claim 4, wherein Generating the scaling factor based on the average signal to interference ratio comprises modifying the average signal to interference ratio by an offset parameter. . The method of claim 5, Generating the average signal to interference ratio includes: Calculating a signal correlation matrix based on the channel estimate; And And calculating the average signal-to-interference ratio based on the signal correlation matrix and the noise statistic. The method of claim 6, Generating the scaling factor based on the average signal to interference ratio is given by the following equation:

Generating the scaling factor in accordance with the present invention, wherein

Is an average signal-to-interference ratio and α represents the offset parameter. The method of claim 7, wherein The offset parameter is obtained from at least one of a plurality of paths of the plurality of path channels processed by the wireless receiver and a plurality of despread values processed by the wireless receiver to generate the channel estimates. Bias removal method from the starting signal to interference ratio caused by The method of claim 6, Computing the average signal to interference ratio based on the signal correlation matrix and the noise statistics: Calculating a weighting factor based on the channel estimate; Calculating the average signal power based on the signal correlation matrix and the weighting factor; Calculating an average noise power based on the noise statistics and the weighting factor; And Calculating the average signal-to-interference ratio based on the average signal power and the average noise power. The method of claim 9, Generating the scaling factor based on the average signal to interference ratio is given by the following equation:

Generating the scaling factor according to

Represents an average signal-to-interference ratio, and α represents the offset parameter. The method of claim 10, The offset parameter is obtained from at least one of a plurality of paths of the plurality of path channels processed by the wireless receiver and a plurality of despread values processed by the wireless receiver to generate the channel estimates. Bias removal method from the starting signal to interference ratio caused by The method of claim 6, Calculating a channel estimation matrix based on the noise statistics and the signal correlation matrix; And Calculating another average signal-to-interference ratio based on the channel estimation matrix, wherein generating the scaling factor based on the average signal-to-interference ratio processes both average signal-to-interference ratios to calculate the scaling factor. And biasing the starting signal-to-interference ratio generated by the wireless receiver. The method of claim 12, Processing both the average signal-to-interference ratio to generate the scaling factor involves the following equation:

Generating the scaling factor according to

Denotes the average signal-to-interference ratio,

Denotes the other signal-to-interference ratio, and α denotes the offset parameter. The method of claim 13, Bias offset is derived from at least one of a plurality of despreading values processed by the wireless receiver to generate the channel estimate and the channel estimation matrix; Way. The method of claim 1, Calculating the starting signal to interference ratio based on the received signal comprises calculating the starting signal to interference ratio based on channel estimates and noise statistics obtained from the received signal. Bias removal method from the starting signal-to-interference ratio generated by the method. The method of claim 15, Computing the starting signal to interference ratio based on the channel estimate and the noise statistics: Calculating a weighting factor based on the channel estimate; Generating a signal power estimate based on the weighting factor; Generating a noise power estimate based on the weighting factor; And And calculating the initiation signal-to-interference ratio based on the generated signal and noise power estimates. The method of claim 15, Calculating the initiation signal-to-interference ratio based on the channel estimate and the noise statistics comprises combining the channel estimate and the noise statistics in an inverse secondary computer to calculate the initiation signal-to-interference ratio. A bias cancellation method from a start signal to interference ratio generated by a wireless receiver. The method of claim 17, Calculating a signal correlation matrix based on the channel estimate; Calculating a channel estimation matrix based on the signal correlation matrix and the noise characteristic; And Calculating a modified channel estimate based on the channel estimation matrix, wherein the combining of the channel estimate and the noise statistics in the inverse secondary computer comprises the inverse two to calculate the initiation signal to interference ratio. Combining the modified channel estimate and the noise statistics in a secondary computer. The method of claim 1, Generating the average signal-to-interference ratio based on the received signal comprises generating the average signal-to-interference ratio. The method of claim 1, Generating the average signal-to-interference ratio comprises smoothing a previous signal-to-interference value associated with the wireless receiver. The method of claim 1, Generating the average signal-to-interference ratio includes identifying a target signal-to-interference ratio as the average signal-to-interference ratio. A signal to interference ratio processor in a wireless receiver that removes a bias from an initiating signal to interference ratio, the method comprising: A start signal-to-interference ratio calculator for calculating the start signal-to-interference ratio based on the signal received by the wireless receiver; An average signal to interference ratio calculator for generating an average signal to interference ratio; An average signal to interference ratio calculator for generating an average signal to interference ratio; And And a bias canceller for canceling said bias at said starting signal-to-interference ratio using said average signal-to-interference ratio. The method of claim 22, The signal-to-interference ratio processor in the wireless receiver for removing bias from the starting signal-to-interference ratio, wherein the average signal-to-interference ratio calculator generates the average signal-to-interference ratio based on channel estimates and noise statistics obtained from the received signal . The method of claim 23, wherein And the average signal-to-interference ratio calculator to obtain the channel estimate and the noise statistics from the despread value obtained from the received signal. The method of claim 23, wherein The bias remover is: A converter for generating a scaling factor based on the average signal to interference ratio; And And a multiplier for removing the bias by increasing the starting signal to interference ratio by the scaling factor. The method of claim 25, And the converter generates the scaling factor by modifying the average signal to interference ratio by an offset parameter. The method of claim 26, The average signal to interference ratio calculator is: A signal statistics estimator for estimating a signal correlation matrix based on the channel estimate; And And a mean signal to interference ratio estimator for calculating the average signal to interference ratio based on the signal correlation matrix and the noise statistics. . The method of claim 27, Wherein the offset parameter is obtained from at least one of a plurality of paths of a plurality of path channels processed by the wireless receiver and a plurality of despread values processed by the wireless receiver. Signal to interference ratio processor in wireless receiver. The method of claim 27, Wherein the average signal to interference ratio estimator comprises one or more secondary computers to calculate an average signal power based on the channel estimate and to calculate an average noise power based on the noise statistics. Signal-to-interference ratio processor in the wireless receiver to eliminate the. The method of claim 29, Wherein the offset parameter is obtained from one of a plurality of paths of a plurality of path channels processed by the wireless receiver and one of a plurality of despread values processed by the wireless receiver. Signal to interference ratio processor at the receiver. The method of claim 27, The average signal-to-interference ratio estimator further comprises a matrix multiplier to square the channel estimation matrix obtained from the signal correlation matrix and the noise statistics, wherein the average signal-to-interference ratio estimator is based on several squared channel estimation matrices. A signal-to-interference ratio processor in a wireless receiver that removes bias from the starting signal-to-interference ratio, characterized by estimating the signal-to-interference ratio. DLt according to claim 31, And the offset parameter is obtained from at least one of a plurality of despread values processed by the wireless receiver and the channel estimation matrix. The method of claim 22, The signal-to-interference ratio processor in the wireless receiver that removes the bias from the starting signal-to-interference ratio, wherein the initiation signal-to-interference ratio calculator calculates the initiation signal-to-interference ratio based on channel estimates and noise statistics obtained from the received signal. . The method of claim 33, The start signal to interference ratio calculator is: A weight calculator for calculating a weighting factor based on the channel estimate; One or more power estimators for generating a signal power estimate and a noise power estimate based on the weighting factor; And And a combiner for obtaining the initiation signal-to-interference ratio from the signal power estimate and the noise power estimate. The method of claim 33, And wherein the initiation signal-to-interference ratio calculator comprises an inverse secondary computer to calculate the initiation signal-to-interference ratio based on the channel estimate and the noise statistics. Signal to Interference Ratio Processor. The method of claim 35, wherein The starting signal-to-interference ratio calculator further includes a channel estimation processor to calculate a modified channel estimate based on the channel estimate, wherein the inverse secondary computer is configured to generate the channel estimate based on the modified channel estimate and the noise statistics. A signal-to-interference ratio processor in a wireless receiver that removes a bias from the initiation signal-to-interference ratio, characterized by generating an initiation signal to interference ratio. The method of claim 36, The channel estimation processor: A channel estimation matrix calculator for calculating a channel estimation matrix based on the noise statistics; And And a matrix multiplier to generate the modified channel estimate based on the channel estimate and the channel estimate matrix. The method of claim 22, And the wireless receiver is located in at least one of a mobile station and a base station. The method of claim 22, And said average signal to interference ratio calculator generates said average signal to interference ratio based on said received signal. The method of claim 22, And the average signal-to-interference ratio calculator smooths a previous signal-to-interference value to generate the average signal-to-interference ratio. The method of claim 22, And the average signal-to-interference ratio calculator identifies a target signal-to-interference ratio as the average signal-to-interference ratio. A computer readable medium stored in a wireless communication device that stores a set of instructions to remove bias from a signal to interference ratio estimate. Instructions for the set of instructions to calculate an initiation signal to interference ratio based on a signal received by the wireless communication device; Instructions for generating an average signal to interference ratio; And Instructions for using the average signal-to-interference ratio to remove the bias from the starting signal-to-interference ratio stored in a wireless communication device that stores a set of instructions to remove the bias from the signal-to-interference ratio estimate. Computer readable media. The method of claim 42, wherein The command to use the average signal to interference ratio to remove the bias from the starting signal to interference ratio is: Calculating a scaling factor based on the average signal to interference ratio; And Storing the set of instructions to remove the bias from the signal-to-interference ratio estimate by the scaling factor to remove the bias from the starting signal-to-interference ratio. Computer-readable media stored on a wireless communication device. The method of claim 42, wherein And store the set of instructions to remove the bias from the signal to interference ratio estimate, wherein the computer readable media is implemented in at least one of hardware and software. The method of claim 42, wherein And store the set of instructions to remove bias from the signal to interference ratio estimate, wherein the wireless communication device comprises a mobile station. The method of claim 42, wherein And a wireless communication device storing a set of instructions to remove bias from the signal to interference ratio estimate, wherein the wireless communication device comprises a base station.

Images