TWI400913B - Selection of a thresholding parameter for channel estimation - Google Patents

Description

Selection of critical parameters for channel estimation

The present disclosure relates generally to communications and, more particularly, to techniques for deriving channel estimates for one of the communication channels.

In a communication system, a transmitter typically processes (e.g., encodes, interleaves, and symbol maps) traffic data to produce data symbols (which are modulation symbols of the data). For a coherent system, the transmitter uses the data symbol multiplex transmission pilot symbols, processes the multiplexed data and pilot symbols to generate a radio frequency (RF) signal, and transmits the RF signal via the communication channel. The channel uses a channel response to distort the RF signal and further utilizes noise and interference to downgrade the RF signal.

The receiver receives the transmitted RF signal and processes the received RF signal to obtain a sample. For coherent data detection, the receiver estimates the response of the communication channel based on the received pilot and derives a channel estimate. The receiver then performs channel detection (eg, equalization) on the samples using channel estimates to obtain data symbol estimates, which are estimates of the data symbols transmitted by the transmitter. The receiver then processes (e.g., demodulates, deinterleaves, and decodes) the data symbol estimates to obtain decoded data.

The quality of the channel estimate can have a large impact on the performance of the data detection and can affect the quality of the symbol estimation and the correctness of the decoded data. Therefore, there is a need in the art for techniques for deriving a high quality channel estimate in a communication system.

Techniques for deriving a high quality channel estimate are described herein. In accordance with an embodiment of the invention, an apparatus comprising at least one processor and a memory is described. The processor derives a first channel impulse response estimate (CIRE) having a plurality of channel sub-samples. The processor may derive an initial CIRE based on a received pilot and may filter the initial CIRE to obtain a first CIRE. The processor selects a critical parameter value based on at least one criterion that may be related to a channel profile, an operational signal to noise ratio (SNR), an expected channel delay spread, a number of channel subsamplings, and the like. The processor derives a second CIRE by zeroing the selected one of the channel sub-samples in the first cIRE based on the critical parameter value. The processor may determine an average energy of the channel subsampling, derive a threshold based on the average energy and the critical parameter value, and zero the channel subsample having energy less than the threshold. The memory can store a table of critical parameter values for one of the different operating cases. The processor can select one of the stored critical parameter values based on the current operational case.

In accordance with another embodiment, a method of deriving a first CIRE having a plurality of channel sub-samples is provided. A critical parameter value is selected based on at least one criterion. A second CIRE is derived by zeroing a selected one of the plurality of channel subsamples based on the critical parameter value.

According to yet another embodiment, an apparatus is described that includes means for deriving a first CIRE having a plurality of channel sub-samples, means for selecting a critical parameter value based on at least one criterion, and for The critical parameter value and zeroing the selected one of the plurality of channel sub-samples to derive a second CIRE component.

Various aspects and embodiments of the invention are described in further detail below.

The word "exemplary" as used herein means "serving as an example, instance, or illustration." Any embodiment described herein as "exemplary" is not necessarily considered as preferred or advantageous over other embodiments.

The channel estimation techniques described herein can be used in a variety of communication systems, such as coded multidirectional proximity (CDMA) systems, timed multidirectional proximity (TDMA) systems, frequency punctured multidirectional proximity (FDMA) systems, and multiple orthogonal divisions. (OFDM) system, orthogonal frequency division multi-directional proximity (OFDMA) system, single carrier FDMA (SC-FDMA) system, and the like. A CDMA system may implement one or more wireless access technologies (RATs) such as Wide Frequency CDMA (W-CDMA), cdma2000, and the like. Cdma2000 covers the IS-2000, IS-856 and IS-95 standards. The TDMA system can implement a RAT, such as the Global System for Mobile Communications (GSM). These various RATs and standards are known in the art. The OFDM system may be an IEEE 802.11a/g system, a handheld device digital video broadcasting (DVB-H) system, a terrestrial television broadcast integrated digital broadcasting service (ISDB-T) system, or the like. The OFDMA system transmits the modulation symbols in the frequency domain on the orthogonal frequency sub-band by using OFDM. The SC-FDMA system transmits modulation symbols in the time domain over the orthogonal frequency sub-band. For clarity, the following description is directed to techniques for a system having multiple frequency sub-bands, which may be OFDM, OFDMA, or SC-FDMA systems.

1 shows a block diagram of a transmitter 110 and a receiver 150 in a wireless communication system 100. For simplicity, transmitter 110 and receiver 150 are each equipped with a single antenna. For the downlink (or forward link), the transmitter 110 can be a component of the base station and the receiver 150 can be a component of the terminal. For the uplink (or reverse link), the transmitter 110 can be a component of the terminal, and the receiver 150 can be a component of the base station. A base station is typically a fixed station and may also be referred to as a Base Transceiver System (BTS), access point, Node B, or some other terminology. The terminal can be fixed or mobile and can be a wireless device, a mobile phone, a personal digital assistant (PDA), a wireless data card, and the like. The channel estimation techniques described herein can be used for a terminal and a base station.

At transmitter 110, transmit (TX) data processor 112 processes (e.g., encodes, interleaves, and symbol maps) the traffic data and generates the data symbols. Pilot processor 114 generates pilot symbols. As used herein, a data symbol is a modulation symbol of a data, a pilot symbol is a modulation symbol of a pilot, and a modulation symbol is a complex value of a point in a signal constellation (eg, for PSK or QAM), and A symbol is usually a complex value. The modulator 120 multiplexes the transmission of the data symbols and pilot symbols, performs modulation on the multiplexed transmission of data and pilot symbols (e.g., for OFDM or SC-FDMA) and generates transmission symbols. The transmission symbol can be an OFDM symbol or an SC-FDMA symbol and transmitted in one symbol period. Transmission unit (TMTR) 132 processes (e.g., converts to analog, amplify, filter, and upconvert) transmission symbols and produces RF signals transmitted via antenna 134.

At receiver 150, antenna 152 receives the RF signal from transmitter 110 and provides the received signal to a receiving unit (RCVR) 154. The receiving unit 154 conditions (eg, filters, amplifies, downconverts, and digitizes) the received signals and provides input samples. Demodulation transformer 160 performs demodulation on the input samples (e.g., for OFDM or SC-FDMA) to obtain the received symbols. Demodulation transformer 160 provides the received pilot symbols to channel estimator/processor 170 and provides the received data symbols to data detector 172. Channel estimator/processor 170 derives a channel estimate for the wireless channel between transmitter 110 and receiver 150 based on the received pilot symbols. Data detector 172 performs data detection (e.g., equalization or matching filtering) on the received data symbols using channel estimates and provides data symbol estimates, which are estimates of the data symbols transmitted by transmitter 110. The RX data processor 180 processes (e.g., symbol demaps, deinterleaves, and decodes) the data symbol estimates and provides decoded data. In general, processing at the receiver 150 is complementary to processing at the transmitter 110.

Controllers/processors 140 and 190 direct the operation of various processing units at transmitter 110 and receiver 150, respectively. The memories 142 and 192 store the code and data for the transmitter 110 and the receiver 150, respectively.

FIG. 2 shows an exemplary multi-layered frame structure 200 that can be used with system 100. The transmission timeline is split into hyperframes, each of which has a predetermined duration of, for example, about one second. Each hyperframe can include (1) a header field for time division multiplexing (TDM) pilot and additional/control information and (2) one for traffic data and frequency division multiplexing (FDM). Pilot data field. The data field can be divided into multiple (O) and other external frames, each of which can be divided into multiple (F) frames, and each frame can be divided into multiple (T) time slots. For example, each hyperframe can include four external frames (O=4), each external frame can include 32 frames (F=32), and each frame can include 15 time slots. (T=15). If each frame has a duration of 10 milliseconds (ms), which is consistent with W-CDMA, each time slot has a duration of 667 microseconds (μs), and each external frame has a 320mS. The duration, and each hyperframe has a duration of about 1.28 seconds. Hyperframes, external frames, frames and slots can also be referred to as some other terminology.

In an embodiment, different radio technologies can be used for different time slots. For example, W-CDMA can be used for some time slots, while OFDM can be used for other time slots. In general, the system can support any or any combination of radio technologies, and each time slot can use one or more radio technologies. The time slot for OFDM is called an OFDM time slot. The OFDM time slot may carry one or more (N) OFDM symbols and may further include a guard period (GP). For example, an OFDM time slot can carry three OFDM symbols and a guard period, wherein each OFDM symbol has a duration of about 210 μs.

FIG. 3 shows an exemplary sub-band structure 300 that can be used with system 100. The system has a total system bandwidth of BW MHz that is split into multiple (K) orthogonal subbands. K can be any integer value, but is usually a power of two (eg, 128, 256, 512, 1024, etc.) to simplify the conversion between time and frequency. The interval between adjacent sub-bands is BW/K MHz. In a spectrum shaping system, the G sub-bands are not used for transmission, but serve as a protection sub-band to allow the system to meet spectral mask requirements, where typically G > 1. The G guard subbands are typically distributed such that G _L = G/2 guard subbands are at the lower band edge and G _U = G/2 guard subbands are at the higher band edges. The remaining U = K - G sub-bands are available for transmission and are referred to as available sub-bands.

In order to facilitate channel estimation, a pilot may be transmitted on a group of M sub-bands uniformly distributed over the entire system bandwidth. The continuous sub-bands in the group may be spaced apart by S sub-bands, where S = K/M. Some of the sub-bands in the group may be in the G _L lower guard sub-bands and are not available for pilot transmission, and some other sub-bands in the group may be in the G _U higher protection sub-bands It is also not available for pilot transmission. For the example shown in Figure 2, the first Z _L subbands in the group are not used for pilot transmission and are referred to as the return-to-zero pilot sub-band, where the subsequent P sub-bands are used for pilot transmission and are This is called the effective pilot sub-band, and the last Z _U sub-bands in the group are the return-to-zero pilot sub-bands, where M = Z _L + P + Z _U .

In an exemplary design, the system utilizes a subband structure in which K = 1024 total subbands, G _L = 68 lower guard subbands, G _{U for} cyclic prefixes appended to each OFDM symbol = 68 higher guard subbands, U = 888 available subbands, M = 128 pilot subbands, P = 1111 available pilot subbands, and C = 108 wafers. Other values can also be used for these parameters.

2 shows an exemplary frame structure, and FIG. 3 shows an exemplary subband structure. The channel estimates described herein can be used with a variety of frame and subband structures.

For the sake of clarity, the following terms are used in the following description. The vector is represented by a bold and underlined text, where the subscript indicates the length of the vector, for example, h _M represents an M × 1 vector or H _K represents a K × 1 vector, where "×1" in the size is implied and It is omitted for clarity. The matrix is represented by bold and underlined text, where the subscripts indicate the matrix size, for example W _{M × K} represents the M x K matrix. The time domain vector is usually represented by a lowercase text, such as h _K , and the frequency domain vector is usually represented by an uppercase text, such as H _K .

The wireless channel between transmitter 110 and receiver 150 is characterized by a time domain channel impulse response h _K or a corresponding frequency domain channel frequency response H _K . The relationship between the channel impulse response and the channel frequency response can be expressed as a matrix: H _K = W _{K × K} . h _K , and equation (1) Among them, the K×1 vector of the impulse response of the h _K-type radio channel, the K×1 vector of the frequency response of the H _K-type radio channel, the W _{K × K-} based K×K Fourier matrix, A K×K inverse Fourier matrix, and “ ^H ” indicates a total inversion.

Equation (1) indicates that the channel frequency response is a fast Fourier transform or a discrete Fourier transform (FFT/DFT) of the channel impulse response. Equation (2) indicates that the channel impulse response is an inverse FFT or an inverse DFT (IFFT/IDFT) of the channel frequency response. The elements of the Fourier matrix W _{K × K} and the elements in row c can be as follows: , where r=1,...,K and c=1,...,K Equation (3) The "-1" in the index of equation (3) is attributed to the index r and c starting from 1 instead of 0 Start.

Transmitter 110 transmits data and pilot symbols to receiver 150 on the available sub-bands. It can be assumed that the data and the pilot symbols have an average energy E _s or E {| X ( k )| ² }= E _s , where X ( k ) is a symbol transmitted on the sub-band K and E {} represents an expectation operating. For simplicity, the following description assumes that each symbol is transmitted at unit power so that E _s = 1.

The received symbol obtained by the receiver 150 in the OFDM symbol period n can be expressed as: Y _K ( n ) = H _X ( n ). X _K ( n )+ η _K ( n ) Equation (4) where X _K ( n ) is the K sub-band containing the K × 1 vector of the transmitted symbol, Y _K ( n ) is the K sub-band The K × 1 vector containing the received symbols, η _K ( n ) is a K × 1 noise vector of K sub-bands, and "." represents an element-wise product.

Each input of X _K ( n ) may be a data symbol of one of the data subbands, a pilot symbol of one pilot subband, or a zero symbol of a subband (eg, a guard subband). For the sake of simplicity, it can be assumed that the pilot symbol has a complex value of 1+j0 and one is =1 value. In this case, the received pilot symbols are only the noise version of the channel gain in H _K ( n ).

If only P pilot subbands are used for pilot transmission, as shown in FIG. 3, the receiver can form an M×1 vector Y _M ( n ), which contains P received by P active pilot subbands. Z _L + Z _U zero symbols of the pilot symbols and the return-to-zero pilot sub-band. The vector Y _M ( n ) can be expressed as: among them and All zero vectors are used, and Y _P ( n ) is the P × 1 vector of the received pilot symbols of the P active pilot subbands.

A variety of techniques are available for estimating channel impulse responses based on received pilot symbols. These technologies include least squares (LS) technology, minimum mean square error (MMSE) techniques, robust MMSE techniques, and forced zero (ZF) techniques.

A least square channel impulse response estimate (CIRE) can be derived ( n ) is: Where h _M ( n ) is an M × 1 channel impulse response vector of M channel sub-samples, and η _M (n) is an M × 1 noise vector of M pilot sub-bands.

Equation (6) indicates that the least square CIRE can be obtained only by taking the M point IFFT/IDFT of the pilot symbol received in Y _M (n). A forced zero CIRE is equal to the least square CIRE.

Export a MMSE CIRE ( n ) is: Where ψ _{h h} = E {h _M ( n ). ( n )} is a M × M channel covariance matrix, and Λ _{η η} = E { η _M ( n ). ( n )} is a M × M noise covariance matrix.

Can export a strong MMSE CIRE ( n ) is: Equation (8) assumes that the sub-samples in the channel impulse response are uncorrelated and have equal power such that Ψ _{h h} = I _{M × M} . Equation (8) further assumes that the noise η _M ( n ) has a zero mean vector and one is Λ _{η η} = N ₀ . An additional white Gaussian noise (AWGN) of the covariance matrix of I _{K × K} , where N ₀ is the variance of the noise and I _{K × K} is a K × K unit matrix.

The receiver may derive an initial CIRE h' _M ( n ) from the OFDM symbols transmitted in the symbol period for each OFDM symbol period n with pilot transmission based on the received pilot symbols. The receiver can derive h' _M ( n ) by using least squares, MMSE, strong MMSE, or some other technique. Therefore, h' _M ( n ) can be equal to ( n ), ( n ) or ( n ).

The receiver can screen the initial CIREs h' _M ( n ) for different OFDM symbol periods to obtain filtered CIREs with improved quality ( n ). Filtering can be performed in a variety of ways.

In an embodiment, the screening of the "internal" OFDM symbols n bordering the left OFDM symbol n-1 and the right OFDM symbol n+1 may be performed as follows: In equation (9), the filtered CIRE of the current OFDM symbol period _M ( n ) is determined based on the initial CIREs of the previous, current, and subsequent OFDM symbol periods.

In an embodiment, the screening of the "left edge" OFDM symbol n that borders only the right OFDM symbol n+1 may be performed as follows: In equation (10), the filtered CIRE of the current OFDM symbol period ( n ) is determined based on the initial CIREs of the current and subsequent OFDM symbol periods.

In an embodiment, the screening of the "right edge" OFDM symbol n that borders only the left OFDM symbol n-1 may be performed as follows: In equation (11), the filtered CIRE of the current OFDM symbol period ( n ) is determined based on the initial CIREs of the previous and current OFDM symbol periods.

In general, time screening of initial CIREs can be performed on any number of past and/or future OFDM symbols. In addition, time filtering can be performed using a finite impulse response (FIR) filter (eg, as shown in equations (9) through (11)), an infinite impulse response (IIR) filter, or some other type of filter. . The screening may also be adaptive, for example, based on the speed of the receiver, the rate of change of the channel state, the operational SNR, and the like.

Receiver can filter the CIRE _M ( n ) performs a limit to obtain a final CIRE _M ( n ). Screened CIRE _M ( n ) contains M channel subsampling ( n ) to ( n ), sub-sampling for each channel ( n ), where (m = 1, ..., M) has a composite gain determined by the radio channel. The limit reserves the channel subsampling with sufficient energy and discards the weak channel subsampling.

In one aspect, the limit is performed based on a critical parameter and a threshold. In order to derive the threshold, The average channel energy of the M channel subsamplings in _M ( n ) can be calculated as follows: among them ( n ) The mth channel in _M (n) is subsampled, and E _avg ( n ) is the average channel energy of the OFDM symbol period n.

In an embodiment, the threshold is defined based on the average channel energy and the critical parameter, as follows: T _h (n) = P . E _avg ( n ), Equation (13) where P is a critical parameter and T _h (n) is a critical value of the OFDM symbol period n. This critical parameter can also be referred to as a critical constant, a scale factor, and the like. The critical value can also be defined as T _h (n) = P ₁ . E _total ( n ), where E _total ( n ) is the total channel energy and P ₁ = P /M is the modified critical parameter.

In general, the threshold T _h (n) can be a function of any amount. The threshold can be a function of the average channel energy and the critical parameter, for example as shown in equation (13). Alternatively, the threshold may be a function of noise energy, energy of several weak channel subsamplings, strongest channel subsampling energy, and the like.

The receiver can perform the limits of the filtered CIRE as follows: , where m=1,...,M equation (14) where _m ( n ) The mth channel in _M ( n ) is subsampled. In the embodiment shown in equation (14), Each of the M channel sub-samples in _M ( n ) individually performs a limit. Calculate each filtered channel subsampling (N) and the energy with the threshold value T _h (n) are compared. If the energy meets or exceeds the threshold T _h (n) , the final channel is subsampled _m ( n ) is set to the filtered channel subsampling ( n ), otherwise set it to 0.

FIG. 4 illustrates an example of an exemplary channel impulse response estimate 400. The energy of the M filtered channel sub-samples is shown by the vertical lines of sub-sample subscripts 1 to M having different heights. The threshold T _h (n) is shown by dashed line 410. The channel sub-sampling with the energy above line 410 is retained and the weak channel sub-sample with the energy below line 410 is zeroed. As can be seen in Figure 4, increasing the threshold and line 410 (by increasing the critical parameter) can result in more channel sub-sampling being zeroed. Conversely, lowering the threshold (by reducing the critical parameter) and line 410 can result in more channel sub-sampling being preserved.

Figure 4 and the above description are for a limited embodiment. Limits can also be enforced in other ways. For example, channel sub-sampling can be ranked from strongest to weakest. Channel subsampling can then be zeroed, one channel subsampling is processed at a time since the weakest channel subsampling, until a certain percentage of the total energy is discarded, and a certain percentage or number of channel subsamplings are zeroed. The percentage can be determined by the critical parameter P.

Channel subsampling in a screened CIRE ( n ) Execute the limit, as described above. Channel sub-sampling in the initial CIRE can also be performed without screening ( n ) enforce the limit.

The receiver can use the final CIRE _M ( n ) achieves various purposes such as data detection, log likelihood ratio (LLR) calculation, and the like. For example, the receiver can be based on a final CIRE with M channel subsampling _M ( n ) derives a final channel frequency response estimate for all K total subbands _K ( n ). The receiver can then use the final channel frequency response estimate _K ( n ) performs equalization or matching screening on the data symbols received in Y _K ( n ) to obtain data symbol estimates _K ( n ). Receiver can also be used _K ( n ) calculates the LLR as the bit of the data symbol estimate.

Computer simulations were performed for the exemplary OFDM systems shown in Figures 2 and 3, where K = 1024, G = 136, U = 888, M = 128, P = 1111, and C = 108. The simulation corresponds to six different operating cases of two channel models and three combinations of coding rate and modulation mechanism. For each simulated operating case, different critical parameter values produce different performance. These simulations show that the critical parameters have a large impact on the quality and performance of the channel estimates. Table 1 gives the critical parameter values that provide the best performance for the six operating cases simulated.

VEHA and PEDB are two channel distribution models well known in the art. One channel is distributed as a statistical model of the channel impulse response and indicates how the communication channel looks in the time domain. The distribution of a channel depends on speed and the environment.

The results of Table 1 are obtained by turbo coding of a larger data block size and a data block transmitted by 12 time slots of 12 OFDM symbols. Data blocks can also be called packets, frames, and the like. For the dual cluster VEHA model, the first cluster starts at 0 μs, the second cluster starts at 10 μs, the two clusters have equal power, and the transmission pulse is a complete sinc function.

Each combination of coding rate and modulation mechanism requires a specific minimum SNR to achieve a target block error rate (BLER), such as 1% BLER. In Table 1, QPSK is used and the SNR required for a ratio of 0.55 is lower than the SNR required to use 16-QAM with a ratio of 0.41. The SNR required for using 16-QAM and the ratio of 0.41 is lower than the ratio using 16-QAM. The required SNR is 0.55. For a particular modulation mechanism, a higher coding rate corresponds to a higher desired SNR. For a particular coding rate, a higher order modulation mechanism corresponds to a higher desired SNR. Table 1 shows that a higher critical parameter value can provide a better performance with a lower SNR for a particular channel distribution.

Table 1 provides the results of some illustrative operational cases. In general, an operational case is characterized by channel distribution, operational SNR, coding and modulation mechanisms, some other parameters, any of these parameters, or any combination of such parameters. Various operational cases can be simulated to determine the critical parameter values that provide the best performance for such operational cases. Different results can be obtained by different system parameters, channel distribution models and/or assumptions.

The appropriate value for the critical parameter P can be determined in various ways. In one embodiment, critical parameter values that provide good performance for various operational cases can be determined by computer simulations, empirical measurements, etc., and can be stored in a lookup table. Thereafter, the current operational case of the receiver can be determined, for example, based on channel distribution, coding and modulation mechanisms, and/or other parameters suitable for the receiver. The critical parameter values corresponding to the current operating case can be retrieved from the lookup table and used for channel estimation.

In another embodiment, the critical parameter value P is selected based on an expected operational SNR. The operational SNR may be estimated based on the received pilot symbols and/or received data symbols. In general, a smaller critical parameter value can be used for a higher SNR, and a larger critical parameter value can be used for a lower SNR.

In yet another embodiment, the critical parameter value P is selected based on the number of channel sub-samplings in the CIRE. The number of channel sub-samplings may be determined by the number of sub-bands used for pilot transmission, the manner in which channel estimation is performed at the receiver, and possibly other factors.

In yet another embodiment, the critical parameter value P is determined based on a high quality channel estimate. The receiver can obtain the high quality channel estimate based, for example, on a TDM pilot or via some other means. A channel distribution of the receiver can be determined based on the high quality channel estimate, and a critical parameter value can be selected based on the channel distribution.

In an embodiment, a new critical parameter value is selected whenever a higher quality channel estimate is required. For example, if a packet is erroneously decoded, a new critical parameter value can be selected. The new critical parameter value can be obtained as follows: P _new = P _old + Δ P , or P _new = P _old - Δ P , where Equation (15) where P _old is the past/current critical parameter value, P _new The new critical parameter value, and ΔP is the step size, which can be set to 0.25 or some other value.

A new channel estimate can be derived based on the new critical parameter value and used to recover the packet. If the packet is still erroneously decoded by the new channel estimate, another critical parameter value can be selected and used to derive another channel estimate, which can then be used to recover the packet. In general, any number of channel estimates can be derived from different critical parameter values. The new critical parameter value can be selected from both sides of the original critical parameter value in another manner. For example, the new critical parameter value can be set to P _old + Δ P , followed by P _old - Δ P , followed by P _old + 2 Δ P , followed by P _old -2 Δ P and the like. A new critical parameter value can be selected and used until the packet is correctly decoded, the maximum number of values is tried, or some other termination condition is encountered. If the packet is decoded correctly, the critical parameter value resulting in successful decoding can be used for subsequent packets. The selection of new critical parameter values can also be triggered by events other than packet errors.

FIG. 5 shows a block diagram of one embodiment of the channel estimator/processor 170 of FIG. Within channel estimator/processor 170, a pilot demodulator (Demod) 512 removes the modulation on the received pilot symbols and also provides zero symbols for the unused pilot subbands. A CIRE processor 514 derives the initial CIRE of the current symbol period based on the output of the pilot demodulator 512. The CIRE processor 514 can derive the initial based on the least squares technique shown in equation (6), the MMSE technique shown in equation (7), the robust MMSE technique shown in equation (8), or some other technique. CIRE. A filter 516 filters the initial CIRE for different symbol periods, such as shown in equations (9), (10), and (11), and provides a filtered CIRE for the current symbol period.

The controller 190 determines the current operational case and selects an appropriate critical parameter value for the current operational case. The memory 192 can store a lookup table (LUT) of different critical parameter values for different operating cases. A threshold calculation unit 520 derives a threshold value T _h (n) of the current symbol period based on the filtered CIRE and critical parameter values, for example, as shown in equations (12) and (13). A unit 518 performs a limit on the channel subsampling of the filtered CIRE based on the threshold from unit 520 and provides a final CIRE for the current symbol period. If desired, an FFT unit 522 can derive a channel frequency response estimate based on the final CIRE.

FIG. 6 shows an embodiment of a process 600 for performing channel estimation using a limit. An initial CIRE is derived for each symbol period with pilot transmission (block 612). The initial CIRE may be derived based on the received pilot symbols for the active pilot subband and the zero symbols of the pilot subband used for zeroing. The initial CIRE may also be derived based on least squares, MMSE, strong MMSE, forced zero, or some other technique. A filtered CIRE is derived for the current symbol period by screening the initial CIRE for the current, previous, and/or future symbol periods (block 614). A first CIRE having multiple channel sub-samples can be set to the initial CIRE or the filtered CIRE for the current symbol period (block 616).

A critical parameter value is selected based on at least one criterion (block 618). For example, the critical parameter value can be selected based on channel distribution, operational SNR, number of channel sub-samplings, and the like. A threshold is derived based on the first CIRE and critical parameter values (block 620). In an embodiment, the average energy of the channel sub-sampling in the first CIRE is determined, and the threshold is derived based on the average energy and the critical parameter value. A second CIRE is derived (block 622) by zeroing the selected one of the channel subsamplings in the first CIRE based on the threshold. In an embodiment, the channel sub-sample having energy less than the threshold is zeroed to obtain the second CIRE. The second CIRE may also be derived by performing a limit on the channel sub-sampling in other ways.

Next, a determination is made as to whether a modified channel estimate is needed (block 624). If the packet is decoded incorrectly, an improved channel estimate is required. If the answer to block 624 is "yes" and if block 626 does not encounter a termination condition, then a new critical parameter value is selected, for example, by changing the current critical parameter value by ΔP (block 628). Next, the process returns to block 620 to (1) determine a new threshold based on the new critical parameter value and (2) zero the selected one of the channel subsamplings of the first CIRE based on the new threshold. Export a new second CIRE. Blocks 620 through 628 can be executed any number of times until a termination condition is encountered. If block 624 determines that a modified channel estimate is not required, or if block 626 determines that a termination condition is encountered, then the program terminates.

Those skilled in the art will appreciate that a variety of different techniques can be used to represent information and signals. For example, the materials, instructions, commands, information, signals, bits, symbols, and wafers referred to above may be represented by voltages, currents, electromagnetic waves, magnetic fields or magnetic particles, light fields, or optical particles, or any combination thereof.

It will be further appreciated by those skilled in the art that the various illustrative logical blocks, modules, circuits, and calculation steps described in connection with the embodiments disclosed herein may be constructed as electronic hardware, computer software, or a combination of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been described above generally. Whether this functionality is built as hardware or software depends on the specific application and design constraints of the overall system. A skilled artisan can construct the described functionality for each particular application in various ways, but this construction decision should not be construed as causing a departure from the scope of the invention.

The various illustrative logic blocks, modules, and circuits described in connection with the embodiments disclosed herein can be combined with a general purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), and a field programmable program. A gated array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination designed to perform the functions described herein. A general purpose processor can be any microprocessor, but in the alternative, the processor can be any conventional processor, controller, microcontroller, or state machine. A processor can also be constructed as a combination of computing devices, such as a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in the same DSP core, or any other configuration.

The methods or calculus steps described in connection with the embodiments disclosed herein may be directly implemented in hardware, a software module executed by a processor, or a combination of both. The software module can exist in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, scratchpad, hard disk, removable disk, CD-ROM or any form known in the art. In the storage medium. An exemplary storage medium is coupled to the processor such that the processor can read information from the storage medium and write the information to the storage medium. In an alternative approach, the storage medium can be integrated into the processor. The processor and the storage medium may reside in an ASIC. The ASIC can reside in a user terminal. In the alternative, the processor and the storage medium may exist as discrete components in a user terminal.

The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the invention. It will be readily apparent to those skilled in the art that various modifications of the embodiments are possible, and the general principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Therefore, it is intended that the invention not be limited to the embodiments shown herein,

100. . . Wireless communication system

110. . . Transmitter

112. . . Transport (RX) data processor

114. . . Pilot processor

120. . . Modulator

132. . . Transmission unit

134. . . antenna

140. . . Controller/processor

142. . . Memory

150. . . receiver

152. . . antenna

154. . . Receiving unit

160. . . Demodulation transformer

170. . . Channel estimator/processor

172. . . Data detector

180. . . RX data processor

190. . . Controller/processor

192. . . Memory

200. . . Multi-layer frame structure

300. . . Subband structure

400. . . Channel impulse response estimate

410. . . dotted line

512. . . Pilot demodulator

514. . . CIRE processor

516. . . Filter

518. . . Limiting unit

520. . . Threshold calculation unit

522. . . FFT unit

Figure 1 shows a block diagram of a transmitter and a receiver.

Figure 2 shows an exemplary multi-layered frame structure.

Figure 3 shows an exemplary sub-band structure.

Figure 4 illustrates the limits for a channel impulse response estimate.

Figure 5 shows a block diagram of a channel estimator/processor at the receiver.

Figure 6 shows a process for performing channel estimation using a limit.

170. . . Channel estimator/processor

190. . . Controller/processor

192. . . Memory

512. . . Pilot demodulator

514. . . CIRE processor

516. . . Filter

518. . . Limiting unit

520. . . Threshold calculation unit

522. . . FFT unit

Claims (38)

An apparatus for deriving a channel estimate for a communication channel, the apparatus comprising: at least one processor configured to derive a first channel impulse response estimate (CIRE) having a plurality of channel sub-samples, and Deriving a second CIRE by zeroing a predetermined percentage of the plurality of channel sub-samples based on a threshold parameter value; and a memory coupled to the at least one processor. The apparatus of claim 1, wherein the at least one processor derives a threshold based on the plurality of channel sub-samples and the critical parameter value, and derives by zeroing a channel sub-sample having energy less than the threshold The second CIRE. The device of claim 1, wherein the at least one processor determines an average energy of the plurality of channel sub-samples, derives a threshold based on the average energy and the critical parameter value, and by having a threshold value less than the threshold value The channel sub-sample of energy is zeroed to derive the second CIRE. The device of claim 1, wherein the at least one processor determines a channel distribution of the device and selects the critical parameter value based on the channel distribution. The apparatus of claim 1, wherein the at least one processor determines an operational signal to noise ratio (SNR) and selects the critical parameter value based on the operational SNR. The apparatus of claim 1, wherein the at least one processor selects the critical parameter value based on a number of channel sub-samplings of the first CIRE. The device of claim 1, wherein the at least one processor determines whether a packet is erroneously decoded, and if the packet is erroneously decoded, selecting one A new critical parameter value and deriving a new second CIRE by zeroing the selected one of the plurality of channel subsamples based on the new critical parameter value. The device of claim 1, wherein the at least one processor selects a different critical parameter value, and derives a different second CIRE based on the first CIRE and the different critical parameter values until a packet is correctly decoded or encounters a Terminate the condition. The device of claim 1, wherein the memory stores a table of critical parameter values for different operating cases, and wherein the at least one processor selects one of the critical parameter values based on the current operating case. The apparatus of claim 1, wherein the at least one processor obtains the received pilot symbols for the subband of the pilot transmission and derives the first CIRE based on the received pilot symbols. The apparatus of claim 1, wherein the at least one processor obtains the received pilot symbols for the sub-band of the pilot transmission, provides zero symbols for the zeroed pilot sub-band, and based on the received guides The frequency symbols and the zero symbols are used to derive the first CIRE. The apparatus of claim 1, wherein the at least one processor derives the first CIRE based on a least squares, a minimum mean square error (MMSE), a strong MMSE, or a forced zero technique. The apparatus of claim 1, wherein the at least one processor derives an initial CIRE for a plurality of symbol periods based on the received pilot, and derives the first CIRE by filtering the initial CIREs. The device of claim 1, wherein the at least one processor derives a current symbol period, at least one previous symbol period, and at least based on the received pilot. An initial CIRE of a future symbol period, and the first CIRE is derived by screening the initial CIREs. The apparatus of claim 1, wherein the at least one processor derives an initial CIRE of the current symbol period and at least one previous symbol period based on the received pilot, and derives the first CIRE by filtering the initial CIREs. The apparatus of claim 1, wherein the at least one processor derives an initial CIRE of the current symbol period and at least one future symbol period based on the received pilot, and derives the first CIRE by filtering the initial CIREs. The device of claim 4, wherein the channel distribution further comprises at least one signal power value corresponding to each signal delay value. The device of claim 9, wherein the current operating case comprises a channel distribution. The device of claim 18, wherein the channel distribution further comprises at least one signal delay and at least one signal power. A method for deriving a channel estimate for a communication channel, the method comprising: deriving a first channel impulse response estimate (CIRE) having a plurality of channel sub-samples; selecting a critical parameter value based on at least one criterion; Deriving a second CIRE by zeroing a predetermined percentage of the plurality of channel sub-samples based on the critical parameter value; wherein selecting the critical parameter value comprises selecting based on a channel distribution including at least one signal delay value of a device The critical parameter value. The method of claim 20, further comprising: Determining an average energy of one of the plurality of channel sub-samples; and deriving a threshold based on the average energy and the critical parameter value, and wherein deriving the second CIRE comprises using a channel having an energy less than the threshold The sample is zeroed to derive the second CIRE. The method of claim 20, wherein the selecting the critical parameter value comprises selecting the critical parameter value based on a channel distribution, an operational signal ratio (SNR), or a number of channel subsamplings of the first CIRE. The method of claim 20, further comprising: determining whether a packet is erroneously decoded; and if the packet is erroneously decoded, selecting a new critical parameter value and by arranging the plurality based on the new critical parameter value The selected one of the channel sub-samples is zeroed to derive a new second CIRE. The method of claim 20, further comprising: deriving an initial CIRE for the plurality of symbol periods based on the received pilot, and wherein deriving the first CIRE comprises screening the initial CIRE to obtain the first CIRE. The method of claim 20, wherein the channel distribution further comprises at least one signal power value corresponding to each signal delay value. An apparatus for deriving a channel estimate for a communication channel, the apparatus comprising: means for deriving a first channel impulse response estimate (CIRE) having a plurality of channel sub-samples; for selecting one based on at least one criterion a component of a critical parameter value; and Means for deriving a second CIRE by zeroing a predetermined percentage of the plurality of channel sub-samples based on the critical parameter value; wherein the means for selecting the critical parameter value is included for A channel distribution of at least one signal delay value to select a component of the critical parameter value. The apparatus of claim 26, further comprising: means for determining an average energy of the plurality of channel sub-samples; and means for deriving a threshold based on the average energy and the critical parameter value, and wherein The means for deriving the second CIRE includes means for deriving the second CIRE by zeroing the channel sub-sample having energy less than the threshold. The apparatus of claim 26, wherein the means for selecting the critical parameter value comprises selecting the critical parameter value based on a channel distribution, an operational signal ratio (SNR), or a number of channel subsamplings of the first CIRE member. The apparatus of claim 26, further comprising: means for determining whether a packet is erroneously decoded; means for selecting a new critical parameter value if the packet is erroneously decoded; and for if the packet is incorrect During decoding, a new second CIRE component is derived by zeroing the selected one of the plurality of channel subsamples based on the new critical parameter value. The apparatus of claim 26, further comprising: means for deriving an initial CIRE of the plurality of symbol periods based on the received pilot, and wherein the component package for deriving the first CIRE A means for screening the initial CIRE to obtain the first CIRE. The device of claim 26, wherein the channel distribution further comprises at least one signal power value corresponding to each signal delay value. An apparatus for deriving a channel estimate for a communication channel, the apparatus comprising: at least one processor configured to derive a first channel impulse response estimate (CIRE) having a plurality of channel sub-samples, based on at least a criterion for selecting a critical parameter value, wherein the at least one criterion comprises a coding rate of the received data, and deriving a second by zeroing the selected one of the plurality of channel sub-samples based on the critical parameter value CIRE; and a memory coupled to the at least one processor. The apparatus of claim 32, wherein the critical parameter value changes inversely to the coding rate. An apparatus for deriving a channel estimate for a communication channel, the apparatus comprising: at least one processor configured to derive a first channel impulse response estimate (CIRE) having a plurality of channel sub-samples, based on at least a criterion for selecting a critical parameter value, wherein the at least one criterion comprises a modulation mechanism of the received data, and deriving a first by zeroing the selected one of the plurality of channel sub-samples based on the critical parameter value a second CIRE; and a memory coupled to the at least one processor. The apparatus of claim 34, wherein the critical parameter value is changed inversely to a cluster point of the number of one of the modulation mechanisms. A non-transitory machine readable medium storing a code, the code When executed by a processor, includes: instructions for deriving a first channel impulse response estimate (CIRE) having a plurality of channel subsamplings; instructions for selecting a critical parameter value based on at least one criterion; Deriving a second CIRE instruction by zeroing a percentage of the plurality of channel sub-samples based on the critical parameter value; wherein selecting the critical parameter value comprises channel distribution based on at least one signal delay value comprising a device Select the critical parameter value. The non-transitory machine readable medium of claim 36, wherein the channel distribution further comprises at least one signal power value corresponding to each signal delay value. An apparatus for deriving a channel estimate for a communication channel, the apparatus comprising: at least one processor configured to derive a first channel impulse response estimate for a channel having a plurality of channel subsamplings ( CIRE), and deriving a second CIRE by zeroing a percentage of the plurality of channel sub-samples based on a threshold, the threshold being derived based on one of the channel operating signal-to-noise ratios (SNR); The memory is coupled to the at least one processor.