Iterative interference cancellation receiver

background

(See Simultaneously Pending Patent Application) This patent application has been assigned to the assignee of this application and is expressly incorporated herein by reference, filed on February 27, 2008, Atoni Docket No. 071339. Coherent SINGLE ANTENNA INTERFERENCE CANCELLATION FOR GSM / GPRS / EDGE, entitled "COHERENT SINGLE ANTENNA INTERFERENCE CANCELLATION FOR GSM / GPRS / EDGE" Related to US Patent Application No. 12 / 038,724.

This patent application was filed on August 19, 2008 and has Atoni Dockett No. 072177, which has been transferred to the assignee of this application and is expressly incorporated herein by reference, "Channel Input Beamforming". Related to the co-pending U.S. Patent Application No. 12 / 193,995, entitled "ENHANCED SINGLE ANTENNA INTERFERENCE CANCELLATION (ESAIC) USING CHANNEL INPUT BEAMFORMING".

The present invention relates to wireless communication, and more particularly to coherent single antenna interference cancellation.

(Background) In many communication systems using GSM, GPRS, EDGE, FDMA, TDMA, CDMA, etc., the receiver's ability to correctly decode the received signal depends on the receiver's ability to maintain carrier synchronization. Dependent. However, as wireless communications become more widespread, increased interference can adversely affect the receiver's ability to maintain this timing.

According to one aspect of the present technology, a method of suppressing interference in wireless communication is an interference suppression filter (an) having a step of receiving a burst of symbols and a first plurality of weights. The filtering burst of symbols is used to perform a filtering step and the filtered burst of symbols to generate data corresponding to the burst of the symbols. Decoding step and a re-encoded burst of A step of encoding the data to generate symbols), a step of calculating a second plurality of weights for the interference suppression filter based on the burst of the re-encoded symbols, and the second. The interference suppression filter having the plurality of weights of the above includes a step of filtering the burst of the re-encoded symbol and a step of decoding the burst of the filtered re-encoded symbol.

According to another aspect of the subject technology, the receiver is configured to filter the burst of the symbol with an antenna configured to receive the burst of the symbol and a first plurality of weights. A filter, a decoder configured to decode the burst of the filtered symbol to generate data corresponding to the burst of the symbol, and the data to generate the burst of the recoded symbol. A encoder configured to encode a plurality of weights, a processor configured to calculate a second plurality of weights for the interference suppression filter based on the burst of the re-encoded symbol, and the re-encoding. The interference suppression filter configured to filter the burst of symbols by the second plurality of weights and the decoder configured to decode the burst of the filtered recoded symbols. Be prepared.

According to yet another aspect of the subject technology, the receiver comprises means for receiving the burst of the symbol, interference suppression means for filtering the burst of the symbol with a first plurality of weights, and said symbol. Decoding means for decoding bursts of the filtered symbols to generate data corresponding to the bursts of, and encoding the data to generate bursts of recoded symbols. With the means for calculating the second plurality of weights for the interference suppressing means based on the burst of the re-encoded symbol, and the means for calculating the second plurality of weights, and the re-encoded symbol with the second plurality of weights. The interference suppressing means for filtering the burst of the filter and the decoding means for decoding the burst of the filtered re-encoded symbol are provided.

According to yet another aspect of the subject technology, the machine-readable medium comprises instructions for suppressing interference in wireless communication. The instruction receives the burst of a symbol and uses a first plurality of weighted interference suppression filters to filter the burst of the symbol and generate the data corresponding to the burst of the symbol. To decode the burst of the symbol and generate the burst of the re-encoded symbol, the data is encoded and a second plurality of weights are applied to the interference suppression filter based on the burst of the re-encoded symbol. Code for calculating and using the interference suppression filter having the second plurality of weights to filter the burst of the recoded symbol and to decode the burst of the filtered recoded symbol. To be equipped with.

It will be appreciated by those skilled in the art that other configurations of the subject art will be readily apparent to those skilled in the art from the following detailed description, in which the various configurations of the subject technology are shown and explained by way of example. As will be appreciated, the subject matter can be configured in other and different ways, and some details of the subject matter may be modified in various other respects, all without departing from the scope of the subject matter. It is possible. Therefore, drawings and detailed descriptions should be considered exemplary in nature and should not be considered limiting.

<figref num="1">The figure which shows the exemplary frame format and burst format in GSM by one aspect of this subject technology.</figref><figref num="2">The figure which shows the receiver for use in a wireless communication system by one aspect of this subject technology.</figref><figref num="3">FIG. 5 shows a subset of symbols, including a first mid-amble symbol selected by the receiver, according to one aspect of the subject technology.</figref><figref num="4">The figure which shows the method of suppressing the interference by one aspect of this subject technology.</figref><figref num="5">The figure which shows the receiver for use in a wireless communication system by one aspect of this subject technology.</figref><figref num="6">A block diagram showing a computer system in which some aspects of the subject technology can be implemented.</figref>

Detailed explanation

Figure 1 shows an exemplary frame format and burst format in GSM. The timeline for downlink transmission is divided into multiple frames. For the traffic channel used to send user-specific data, each multiframe, such as the exemplary multiframe 101, contains 26 TDMA frames labeled as TDMA frames 0-25. Traffic channels are sent in TDMA frames 0-11 and TDMA frames 13-24 of each multiframe, identified by the letter "T" in FIG. The control channel identified by the letter "C" is sent in TDMA frame 12. No data is sent in idle TDMA frame 25 (identified by the letter "I"), which is used by wireless devices to make measurements of nearby base stations.

Each TDMA frame, such as the exemplary TDMA frame 102, is further subdivided into eight time slots, labeled as time slots 0-7. Each active wireless device / user is assigned one time slot index for the duration of the call. User-specific data for each wireless device is sent in the time slot assigned to that wireless device and in the TDMA frame used for the traffic channel.

Transmission in each time slot is called a "burst" in GSM. Each burst, such as the exemplary burst 103, has two tail fields, two data fields, a training sequence (or midamble) field, and a guard period (GP). )including. The number of bits in each field is shown in parentheses. GSM defines eight different training sequences that can be sent in the training sequence field. Each training sequence, such as Midamble 104, is defined to contain 26 bits, with the first 5 bits being repeated and the second 5 bits being repeated. Each training sequence also has a correlation between that sequence and the 16-bit truncated version of that sequence: (a) 16 if the time shift is zero, (b) ± 1, ± 2, ± 3, ± for the time shift. It is defined to be equal to zero for 4, ± 5, and (3) zero or non-zero for all other time shifts.

A technique for locating mid-ambles in a symbol burst is a continuous comparison of hypotheses about mid-amble positions, which hypotheses are assumed for known mid-amble sequences and symbol bursts. Determine if the highest correlation energy is given to the position. This method is very susceptible to interference from the multipath of the same mid-amble sequence, so that the correlation energy of the inaccurate hypothesis can be affected by the time-delayed copy of the mid-amble sequence.

FIG. 2 shows a receiver for use in a wireless communication system according to one aspect of the subject technology. The receiver 200 includes an antenna 210 configured to receive wireless signals. The receiver 200 can be used in various communication systems, but for the sake of clarity, the receiver 200 is specifically described herein with respect to the GSM system. The received signal is given to the preprocessor 220, which demodulates the signal and produces a received sample. The pre-processor 220 may include a GMSK-BPSK rotator that performs phase rotation on the received sample. The timing estimator 230 receives a sample from the preprocessor 220, makes some hypotheses about where the training symbol sequence (ie, the midamble) begins during a data burst, and gives some hypothetical channel estimates. .. Interference suppressor 240 performs single antenna interference cancellation on each hypothesized channel, and midamble estimator 250 midamble estimation for each hypothesis. error) is generated. The timing determination circuit 260 compares the mid-amble estimation error for each hypothesis and selects the hypothesis with the lowest mid-amble estimation error. The hypothesis selection by the timing circuit 260 represents the position of the symbol in the burst, where the midamble is presumed to begin. Using the selected timing hypothesis, the received sample is given to the interference suppressor 240, which recalculates the training weights based on the entire burst of the received sample ( re-calculates), filtering the entire burst. The filtered signal is then given to the data processor 270, which decodes the received symbol based on the selected timing hypothesis and outputs the data corresponding to the received symbol. The decoding process is the decoding error (decoding). Any one of a plurality of error correction methods known to those of skill in the art can be used to reduce error) and to provide accurate decrypted data. The decoded data is given to the encoder 280, which re-encodes the data to give a burst of the re-encoded symbols, which are of the re-encoded symbols by the interference suppressor 240. It is used to recalculate the training weights based on the burst, and then the encoder 280 uses the recalculated weights to filter the burst of recoded symbols. The filtered signal is then given to the data processor 270, which decodes the received symbol based on the selected timing hypothesis and outputs the data corresponding to the received symbol.

Accurate time synchronization is achieved non-coherently (eg, by selecting the maximum correlated energy sum) or coherently by performing interference suppression to give an estimate of the making up symbol. This symbol estimate can be compared to a symbol known prior to the training sequence to determine its estimation error.

To begin the search for the first midamble symbol, the timing estimator 230 opens a "window" near the estimated beginning of the midamble sequence. The position of the first symbol in the mid-amble sequence can be estimated for a given burst based on the known structure of each burst. For example, as shown in FIG. 1, the beginning of the mid amble 104 in burst 103 begins at bit 62 of the burst. Based on this known structure, the timing estimator 230 selects a window 105 consisting of bits representing a set of hypotheses about where the first midamble symbol can be located. An exemplary window 105 is shown in more detail in FIG.

As can be seen with reference to FIG. 3, the exemplary window 105 comprises 11 symbols labeled with Δ = 0 to Δ = 10. Each Δ value represents the position of the symbol in the window. However, referring to the position of the symbol in the entire burst, the Δ value is offset by the offset value (for example, Δ = 5 can be offset by 61 to represent the position of this symbol in the entire burst). For the first seven symbols in window 105, the timing estimator 230 uses a channel estimate from a sequence of five consecutive symbols (representing GSM's five-tap channel format). To generate. For example, the symbol Δ = 0 is the channel estimate.<maths num="1"><img file="JP5512805B2_D0001.tif" /></maths>

Corresponding to, symbol Δ = 1 is the channel estimate<maths num="2"><img file="JP5512805B2_D0002.tif" /></maths>

Corresponds to, and so on. Each of these channel estimates is then processed by the interference suppressor 240 and the midamble estimator 250 to determine the midamble estimation error of the estimate, and the corresponding estimated midamble symbol is determined.

In this exemplary embodiment, the window 105 is shown to consist of exactly 11 symbols, but the scope of the invention is not limited to such configurations. Instead, any window size (up to the size of the entire data burst) can be selected so that it will be immediately apparent to those skilled in the art. For example, according to one aspect of the subject technology, the size of the search window may be chosen to be twice the size of the expected minimum propagation delay. Alternatively, the search window size can be parameterized based on any other reference value known to those of skill in the art.

According to one aspect, the channel estimate<maths num="3"><img file="JP5512805B2_D0003.tif" /></maths>

Is generated by the timing estimator 230 by correlating the received sample (corresponding to the assumed delay) with the reference sample (ie, known mid-amble sequence) for each hypothesis. Correlation R between the received signal y and the mid-amble sequence i for the assumed delay Δ_ysBased on (Δ), the channel estimate can be calculated as follows.<maths num="4"><img file="JP5512805B2_D0004.tif" /></maths>

To test the hypothesis corresponding to each channel estimate, the interference suppressor 240 performs a single antenna interference cancellation (SAIC) for each estimated channel. SAIC oversampled and / or true / imagined the signal so that weights could be applied to the virtual antenna to form a beam in the direction of the desired transmitter and a beam null in the direction of the unwanted interferer. Using the above decomposition, we can give a virtual antenna with a separate sample sequence. In general, SAIC can be achieved with one or more real antennas at the receiver by using space-time processing, where "space" is with in-phase components. It can be virtually achieved with inphase and quadrature components, and "time" can be achieved with late and early samples.

For example, a set of spatial and temporal samples at time k = 1 ... M x₁Given (k), here,<maths num="5"><img file="JP5512805B2_D0005.tif" /></maths>

And s_kIs a mid-amble / pseudo-mid-amble signal at time k,<maths num="6"><img file="JP5512805B2_D0006.tif" /></maths>

Is (ν + 1) × 1 mid amble / pseudo mid amble vector,<maths num="7"><img file="JP5512805B2_D0007.tif" /></maths>

Is an M × 1 received midamble / pseudo midamble vector, and a set of spatial and temporal samples is<maths num="8"><img file="JP5512805B2_D0008.tif" /></maths>

Can be defined as, however, X_kIs an M × (L + 1) × 1 vector of a spatiotemporal sample with a space length of M and a time length of L + 1. Therefore, the space / time structure matrix is<maths num="9"><img file="JP5512805B2_D0009.tif" /></maths>

However, [X] is an M (L + 1) × p-ν matrix, and p is the length of a mid- or pseudo-mid-amble (data-assisted type). ).

Therefore, [X] and<maths num="10"><img file="JP5512805B2_D0010.tif" /></maths>

Given, the suppression filter W_SAICBy estimating the reference symbol sequence at the channel output, it can be calculated as follows according to one aspect of the present disclosure.<maths num="11"><img file="JP5512805B2_D0011.tif" /></maths>

However, W = (ν-1) × M (L + 1),<maths num="12"><img file="JP5512805B2_D0012.tif" /></maths>

Is<maths num="13"><img file="JP5512805B2_D0013.tif" /></maths>

Or<maths num="14"><img file="JP5512805B2_D0014.tif" /></maths>

Equal to, and here<maths num="15"><img file="JP5512805B2_D0015.tif" /></maths>

Is.

The above formula is<maths num="16"><img file="JP5512805B2_D0016.tif" /></maths>

Can be rewritten as.

The output of the interference suppressor 240 is<maths num="17"><img file="JP5512805B2_D0017.tif" /></maths>

In the form of, here<maths num="18"><img file="JP5512805B2_D0018.tif" /></maths>

Represents an estimate of the midamble sequence. The difference between the estimated mid-amble sequence and the known mid-amble sequence is given in Equation 6 below.<maths num="19"><img file="JP5512805B2_D0019.tif" /></maths>

Determined by each time t_iMidamble estimation error about e_m(t_i) Is acquired. Each time t_iIs the assumed position Δ_iTo offset T from the start of the burst_sIs equal to the sum of.<maths num="20"><img file="JP5512805B2_D0020.tif" /></maths>

Each time t_iMidamble estimation error e_m(t_i) Is determined, the timing determination block 260 will determine which hypothesis has the lowest estimation error e_mIs determined, and other assumed timing values are discarded.

Using the selected timing hypothesis, the received sample is given to the interference suppressor 240, which recalculates the training weights based on the entire burst of the received sample and filters the entire burst. The filtered signal is then given to the data processor 270, which decodes the received symbol based on the selected timing hypothesis and outputs the data corresponding to the received symbol. The decoding process can use any one of a plurality of error correction methods known to those of skill in the art to reduce the decoding error number and to provide accurate decrypted data. The decoded data is given to the encoder 280, which re-encodes the data to give a burst of the re-encoded symbols, which are of the re-encoded symbols by the interference suppressor 240. It is used to recalculate the training weights based on the burst, and then the encoder 280 uses the recalculated weights to filter the burst of recoded symbols. The filtered signal is then given to the data processor 270, which decodes the received symbol based on the selected timing hypothesis and outputs the data corresponding to the received symbol.

For example, according to one aspect of the disclosure, a full rate channel (eg, TCH / FS as defined by the 3GPP standard) is a 260-bit (eg, 182 class 1 bits and 78 class 2 bits) packets. Can have a size. Class 1 bits are convolutional coded at code rate = 1/2. Prior to coding, CRC can be applied to 50 bits of class 1 bits (known as class 1a), which are then cyclic degenerate codes and polynomial generators g (D) = 1 + D + D.³Is convolutional coded using. Following coding and subsequent puncturing (channel dependent), the 456-bit message block is interleaved over eight consecutive frames, then modulated and transmitted. The repetition between message decoding and burst equalization has the advantage of interleaving (which evenly distributes the error and makes the error correction channel code work best) and the advantage of channel coding present in class 1 bits. And are obtained together, which gives a better reference sample for blind interference cancellation, as described in more detail above.

According to one aspect of the disclosure, the interference suppression filter used to filter the burst of recoded symbols uses a higher-order model than that used to filter the burst of symbols in the first iteration. can do. This is because the reduction in signal noise achieved by decoding the pre-filtered signal (with error correction) improves curve fitting for higher-order models (eg, models with an increasing time dimension L). ) Can be used.

According to one aspect of the present disclosure, the data processor 270 comprises a soft output generator that receives a signal from the timing determination block 260 and generates a soft decision indicating confidence in the detection bits. Soft output generators can implement the Ono algorithm, as is known to those of skill in the art. The data processor 270 may further be equipped with a de-interleaver that de-interleaves the soft verdicts and passes the soft verdicts to the Viterbi decoder, where the Viterbi decoder deinterleaves the soft verdicts. Is decoded and the decrypted data is output.

FIG. 4 is a flowchart showing a method for suppressing interference according to one aspect of the present subject technology. This method begins at step 401, where a burst of symbols is received. At step 402, multiple timing hypotheses are generated for the burst of symbols. At step 403, the receiver calculates multiple weights for the interference suppression filter for each timing hypothesis based on a subset of the symbol bursts. For each timing hypothesis, a subset of symbol bursts is filtered at step 404 by a corresponding first plurality of weighted interference suppression filters. At step 405, selection One of multiple timing hypotheses corresponding to criteria) is selected. The selection criterion may be, for example, a mid-amble estimation error. At step 406, the symbol burst is filtered using a first multi-weight interference suppression filter. At step 407, the filtered symbol burst is decrypted to generate the data corresponding to the symbol burst. The decoder can implement any one of multiple error correction procedures known to those of skill in the art. At step 408, the decrypted data is re-encoded to generate a burst of re-encoded symbols, which is then iteratively processed in steps 409-411. First, a second plurality of weights are calculated for the interference suppression filter in step 409 based on the burst of recoded symbols. Then, in step 410, the burst of recoded symbols is filtered using a second multi-weight interference suppression filter. At step 411, the burst of filtered recoded symbols is decrypted again.

FIG. 5 shows a receiver for use in a wireless communication system according to one aspect of the subject technology. The receiver 500 includes, for example, an antenna module 510 configured to receive wireless signals such as RF modulated GSM signals. The received signal is given to a pre-processor module 520 that demodulates the signal and produces a received sample. The preprocessor module 520 may also include a GMSK-BPSK rotator that performs phase rotation on the received sample. The timing estimation module 530 is a pre-processing module (pre-processing). It receives a sample from module) 520 and makes some hypotheses about where the training symbol sequence (midamble) begins during a data burst and gives some hypothetical channel estimates. The interference suppression module 540 performs single antenna interference cancellation by calculating a plurality of filter weights for each hypothesis and then applying a filter with the calculated weights to each channel estimation hypothesis. The mid-amble estimation module 550 generates a mid-amble estimation error for each hypothesis. The timing determination module 560 compares the midamble estimation error for each hypothesis and selects the hypothesis with the lowest midamble estimation error. The hypothesis selection by the timing determination module 560 represents the position of the symbol in the burst, where the midamble is presumed to begin. Using the selected timing hypothesis, the received sample is given to the interference suppression module 540, which recalculates the training weights based on the entire burst of the received sample and filters the entire burst. The filtered signal is then given to the data processor module 570, which processes the received symbol based on the selected timing hypothesis and outputs the data corresponding to the received symbol.

In the above exemplary embodiment, the interference suppression filter is described as a single antenna interference canceling filter, but the scope of the present invention is not limited to such an embodiment. Instead, as will be apparent to those skilled in the art, the subject technology also applies to systems with multiple antennas, such systems that, for example, dual antenna interference cancellation (DAIC), or to those skilled in the art. Can be performed by any other known multiple antenna interference canceling method.

FIG. 6 is a block diagram showing a computer system 600 in which certain aspects can be implemented. The computer system 600 includes a bus 602 or other communication mechanism for transmitting information and a processor 604 combined with the bus 602 for processing information. Computer system 600 also includes memory 606, such as random access memory (RAM) or other dynamic storage device, coupled to bus 602, which stores information and instructions to be executed by processor 604. Memory 606 can also be used to store time-variable or other intermediate information during the execution of instructions to be executed by processor 604. Computer system 600 further includes a data storage device 610, such as a magnetic disk or optical disk, coupled to bus 602 that stores information and instructions.

The computer system 600 may be coupled via an I / O module 608 to a display device (not shown) such as a cathode ray tube (CRT) or liquid crystal display (LCD) that displays information to the computer user. For example, input devices such as keyboards and mice that convey information and command selection to processor 604 may also be coupled to computer system 600 via I / O module 608.

According to one aspect, interference suppression is performed by computer system 600 in response to processor 604 executing one or more sequences of one or more instructions contained in memory 606. Such instructions may be read into memory 606 from another machine-readable medium, such as the data storage device 610. Upon execution of a sequence of instructions contained in main memory 606, processor 604 executes the process steps described herein. One or more processors in a multi-processing configuration may also be used to execute a sequence of instructions contained in memory 606. In alternative embodiments, hard-wired circuit configurations may be used in place of or in combination with software instructions to implement various embodiments. Therefore, the embodiment is not limited to any specific combination of hardware circuit configuration and software.

As used herein, "machine-readable". The term "medium)" refers to any medium involved in giving instructions to processor 604 for execution. Such media can take many forms, including but not limited to non-volatile media, volatile media, and transmission media. Non-volatile media include, for example, optical or magnetic disks such as data storage device 610. Volatile media include dynamic memory such as memory 606. Transmission media include coaxial cables, copper wires, and optical fibers, including wires with bus 602. The transmission medium can also be in the form of acoustic or light waves, such as those produced during radio frequency and infrared data communication. Common forms of machine-readable media include, for example, floppy (registered trademark) disks, flexible disks, optical discs, magnetic tapes, any other magnetic medium, CD-ROM, DVD, any other optical medium, punch cards, paper tape. Includes any other physical medium with a hole pattern, RAM, PROM, EPROM, flash EPROM, any other memory chip or cartridge, carrier, or any other medium that can be read by a computer.

The various exemplary blocks, modules, elements, components, methods, and algorithms described herein can be implemented by those skilled in the art as electronic hardware, computer software, or a combination of both. Will be understood. Furthermore, these can be classified differently from those described. To demonstrate this replacement of hardware and software, various exemplary blocks, modules, elements, components, methods, and algorithms have been generally described above by their functionality. Whether such functionality is implemented as hardware or software depends on the design constraints imposed on a particular application and the entire system. One of ordinary skill in the art can implement the described functionality in various ways for each particular application.

It will be appreciated that the specific order or hierarchy of steps or blocks in the disclosed process provides an exemplary approach. It will be appreciated that the specific order or hierarchy of steps or blocks in a process can be reordered based on design preferences. The attached method claims present the elements of the various steps in a sample order and do not imply that they are limited to a particular order or hierarchy.

The above description is provided to allow any person skilled in the art to implement the various aspects disclosed herein. Various modifications to these embodiments will be readily apparent to those of skill in the art, and the comprehensive principles defined herein can be applied to other embodiments. Therefore, the claims are not intended to be limited to the embodiments presented herein, but should be given the full scope consistent with the language claims. , References to singular elements are not intended to mean "only one" unless otherwise stated as "one and only one", rather "one and only one". Or intended to mean "one or more". Unless otherwise stated, the term "some" may be one or more. points to more). The masculine pronouns (eg, "his") include feminine and neuter (eg, "her" and "it"), and vice versa. All structural and functional equivalents of the elements of the various aspects described throughout this disclosure, which will be known to those of skill in the art or will become known later, are expressly herein by reference. Incorporated in and intended to be incorporated by claim. Moreover, the content disclosed herein is not intended to be made publicly available, whether or not such disclosure is expressly stated in the claims. Claim elements use the phrase "means to do", unless the elements are explicitly stated, or in the case of a method claim, the phrase "step to do". In use, unless the element is described, it should not be construed under the provisions of Article 112, Paragraph 6 of the US Patent Act.<u style="single">Hereinafter, the invention described in [Claims] at the time of filing the invention of the present application will be added.</u><u style="single">[1]</u><u style="single">It is a method of suppressing interference in wireless communication.</u><u style="single">The step of receiving a burst of symbols and</u><u style="single">A step of filtering the burst of the symbol using a first multi-weight interference suppression filter.</u><u style="single">A step of decoding the burst of the filtered symbol and a step of decoding the burst of the filtered symbol in order to generate data corresponding to the burst of the symbol.</u><u style="single">A step of encoding the data to generate a burst of re-encoded symbols,</u><u style="single">A step of calculating a second plurality of weights for the interference suppression filter based on the burst of the recoded symbols,</u><u style="single">A step of filtering the burst of the re-encoded symbol using the interference suppression filter having the second plurality of weights.</u><u style="single">The step of decoding the burst of the filtered, re-encoded symbol,</u><u style="single">How to prepare.</u><u style="single">[2]</u><u style="single">Filtering the burst of the symbol using an interference suppression filter with a first plurality of weights</u><u style="single">To generate multiple timing hypotheses for the burst of the symbol;</u><u style="single">To calculate the corresponding weights for the interference suppression filter for each timing hypothesis based on a subset of the symbols bursts;</u><u style="single">For each timing hypothesis, filtering the subset of bursts of the symbol using the interference suppression filter with the corresponding plurality of weights;</u><u style="single">Selecting one of the plurality of timing hypotheses corresponding to the selection criteria, and the selected timing hypothesis corresponds to the first plurality of weights;</u><u style="single">Using the interference suppression filter having the first plurality of weights to filter the burst of the symbol;</u><u style="single">To prepare</u><u style="single">The method described in [1].</u><u style="single">[3]</u><u style="single">The method according to [2], wherein the selection criterion is a mid-amble estimation error.</u><u style="single">[4]</u><u style="single">The mid-amble estimation error is for each timing hypothesis.</u><u style="single">Determine the estimation channel corresponding to the timing hypothesis</u><u style="single">Perform single antenna interference cancellation for said estimated channel to obtain an estimated midamble sequence, and</u><u style="single">Comparing the estimated midamble sequence with a previously known midamble sequence to determine the midamble estimation error,</u><u style="single">Calculated by</u><u style="single">The method described in [3].</u><u style="single">[5]</u><u style="single">The determination of the estimated channel corresponding to each symbol comprises selecting a predetermined number of adjacent symbols in the subset corresponding to the number of taps in the estimated channel.</u><u style="single">The method described in [4].</u><u style="single">[6]</u><u style="single">The generation of the plurality of timing hypotheses estimates the position of the first midamble symbol in the burst of the symbol and selects the subset of the burst of the symbol from the symbols centered on the estimated position. To do and to prepare</u><u style="single">The method described in [2].</u><u style="single">[7]</u><u style="single">The first plurality of weights for each timing hypothesis are:</u><maths num="21"><img file="JP5512805B2_D0021.tif" /></maths><u style="single">Calculated by solving,</u><u style="single">In addition, it should be noted.</u><maths num="22"><img file="JP5512805B2_D0022.tif" /></maths><u style="single">Is a vector corresponding to an estimate of the subset of symbols, and [X] is a matrix of spatiotemporal samples of bursts of the symbols.</u><u style="single">The method described in [2].</u><u style="single">[8]</u><u style="single">The second plurality of weights</u><maths num="23"><img file="JP5512805B2_D0023.tif" /></maths><u style="single">Calculated by solving,</u><u style="single">In addition, it should be noted.</u><maths num="24"><img file="JP5512805B2_D0024.tif" /></maths><u style="single">Is a vector corresponding to an estimate of said subset of symbols, and [X] is a matrix of spatiotemporal samples of bursts of said symbols.</u><u style="single">The method described in [1].</u><u style="single">[9]</u><u style="single">Decoding the burst of the filtered symbol comprises performing error correction on the burst of the filtered symbol.</u><u style="single">The method described in [1].</u><u style="single">[10]</u><u style="single">The interference suppression filter is a single antenna interference canceling (SAIC) filter or a dual antenna interference canceling (DAIC) filter.</u><u style="single">The method described in [1].</u><u style="single">[11]</u><u style="single">The interference suppression filter uses a higher-order model to filter the burst of the re-encoded symbol than is used to filter the burst of the symbol.</u><u style="single">The method described in [1].</u><u style="single">[12]</u><u style="single">With an antenna configured to receive a burst of symbols,</u><u style="single">An interference suppression filter configured to filter bursts of said symbols with a first plurality of weights.</u><u style="single">A decoder configured to decode the burst of the filtered symbol to generate data corresponding to the burst of the symbol.</u><u style="single">A encoder configured to encode the data to generate a burst of recoded symbols, and</u><u style="single">A processor configured to calculate a second plurality of weights for the interference suppression filter based on the burst of the recoded symbol.</u><u style="single">The interference suppression filter configured to filter the burst of the recoded symbol by the second plurality of weights.</u><u style="single">With the decoder configured to decode the burst of the filtered recoded symbols,</u><u style="single">A receiver equipped with.</u><u style="single">[13]</u><u style="single">With a timing estimator configured to generate multiple timing hypotheses for the burst of the symbol;</u><u style="single">With the processor configured to calculate corresponding weights for each timing hypothesis for the interference suppression filter based on a subset of the bursts of the symbols;</u><u style="single">With the interference suppression filter configured for each timing hypothesis to filter the subset of bursts of the symbol with the corresponding plurality of weights;</u><u style="single">The processor configured to select one of the plurality of timing hypotheses corresponding to the selection criteria, and the selected timing hypothesis corresponds to the first plurality of weights:</u><u style="single">With the interference suppression filter configured to filter the burst of the symbol with the first plurality of weights;</u><u style="single">The receiver described in [12], further equipped with.</u><u style="single">[14]</u><u style="single">The receiver according to [13], wherein the selection criterion is a mid-amble estimation error.</u><u style="single">[15]</u><u style="single">The processor estimates the mid-amble estimation error for each timing hypothesis.</u><u style="single">Determine the estimation channel corresponding to the timing hypothesis</u><u style="single">Perform single antenna interference cancellation for said estimated channel to obtain an estimated midamble sequence, and</u><u style="single">To determine the midamble estimation error, the estimated midamble sequence is compared to a previously known midamble sequence.</u><u style="single">By</u><u style="single">It is configured to calculate,</u><u style="single">The receiver described in [14].</u><u style="single">[16]</u><u style="single">The determination of the estimated channel corresponding to each symbol comprises selecting a predetermined number of adjacent symbols in the subset corresponding to the number of taps in the estimated channel, according to [15]. Receiver.</u><u style="single">[17]</u><u style="single">The timing estimator is</u><u style="single">By estimating the position of the first midamble symbol in the burst of the symbol and selecting the subset of the burst of the symbol from the symbols centered on the estimated position.</u><u style="single">It is configured to generate the plurality of timing hypotheses.</u><u style="single">The receiver described in [13].</u><u style="single">[18]</u><u style="single">The processor</u><maths num="25"><img file="JP5512805B2_D0025.tif" /></maths><u style="single">Is configured to calculate the first plurality of weights for each timing hypothesis.</u><u style="single">In addition, it should be noted.</u><maths num="26"><img file="JP5512805B2_D0026.tif" /></maths><u style="single">Is a vector corresponding to an estimate of said subset of symbols, and [X] is a matrix of spatiotemporal samples of bursts of said symbols.</u><u style="single">The receiver described in [13].</u><u style="single">[19]</u><u style="single">The processor</u><maths num="27"><img file="JP5512805B2_D0027.tif" /></maths><u style="single">Is configured to calculate the second plurality of weights by solving.</u><u style="single">In addition, it should be noted.</u><maths num="28"><img file="JP5512805B2_D0028.tif" /></maths><u style="single">Is a vector corresponding to an estimate of said subset of symbols, and [X] is a matrix of spatiotemporal samples of bursts of said symbols.</u><u style="single">The receiver described in [12].</u><u style="single">[20]</u><u style="single">Decoding the burst of the filtered symbol comprises performing error correction on the burst of the filtered symbol.</u><u style="single">The receiver described in [12].</u><u style="single">[21]</u><u style="single">The interference suppression filter is a single antenna interference canceling (SAIC) filter or a dual antenna interference canceling (DAIC) filter.</u><u style="single">The receiver described in [12].</u><u style="single">[22]</u><u style="single">The interference suppression filter is configured to use a higher order model than used to filter the burst of the symbol to filter the burst of the recoded symbol.</u><u style="single">The receiver described in [12].</u><u style="single">[23]</u><u style="single">Means for receiving a burst of symbols,</u><u style="single">Interference suppression means for filtering the burst of the symbol with the first plurality of weights,</u><u style="single">Decoding means for decoding the burst of the filtered symbol and decoding means for generating the data corresponding to the burst of the symbol.</u><u style="single">Means for encoding the data to generate a burst of recoded symbols, and</u><u style="single">A means for calculating a second plurality of weights for the interference suppression means based on the burst of the recoded symbol, and</u><u style="single">With the interference suppression means for filtering the burst of the recoded symbol by the second plurality of weights.</u><u style="single">With the decoding means for decoding the burst of the filtered re-encoded symbol,</u><u style="single">A receiver equipped with.</u><u style="single">[24]</u><u style="single">With means for generating multiple timing hypotheses for bursts of said symbols;</u><u style="single">A means for calculating a plurality of corresponding weights for the interference suppression means, for each timing hypothesis, based on a subset of the symbol bursts;</u><u style="single">With the interference suppression means for filtering the subset of bursts of the symbol for each timing hypothesis with the corresponding plurality of weights;</u><u style="single">A means for selecting one of the plurality of timing hypotheses corresponding to the selection criteria, and the selected timing hypothesis corresponds to the first plurality of weights;</u><u style="single">With the interference suppression means for filtering the burst of the symbol with the first plurality of weights;</u><u style="single">The receiver according to [23], further equipped with.</u><u style="single">[25]</u><u style="single">The receiver according to [24], wherein the selection criterion is a mid-amble estimation error.</u><u style="single">[26]</u><u style="single">The means for calculating the midamble estimation error for each timing hypothesis is</u><u style="single">Means for determining the estimation channel corresponding to the timing hypothesis and</u><u style="single">Means for performing single antenna interference cancellation on said estimated channel to obtain an estimated midamble sequence, and</u><u style="single">To determine the midamble estimation error, a means for comparing the estimated midamble sequence with a previously known midamble sequence, and</u><u style="single">To prepare</u><u style="single">The receiver described in [25].</u><u style="single">[27]</u><u style="single">The means for determining the estimated channel corresponding to each symbol comprises means for selecting a predetermined number of adjacent symbols in the subset corresponding to the number of taps in the estimated channel.</u><u style="single">The receiver described in [26].</u><u style="single">[28]</u><u style="single">The means for generating the plurality of timing hypotheses is a means for estimating the position of the first midamble symbol in the burst of the symbol, and a burst of the symbol from the symbol centered on the estimated position. With means for selecting said subset of.</u><u style="single">The receiver described in [24].</u><u style="single">[29]</u><u style="single">The means for calculating the first plurality of weights for each timing hypothesis</u><maths num="29"><img file="JP5512805B2_D0029.tif" /></maths><u style="single">With the means to solve</u><maths num="30"><img file="JP5512805B2_D0030.tif" /></maths><u style="single">Is a vector corresponding to an estimate of said subset of symbols, and [X] is a matrix of spatiotemporal samples of bursts of said symbols.</u><u style="single">The receiver described in [24].</u><u style="single">[30]</u><u style="single">The means for calculating the second plurality of weights is</u><maths num="31"><img file="JP5512805B2_D0031.tif" /></maths><u style="single">With the means to solve</u><maths num="32"><img file="JP5512805B2_D0032.tif" /></maths><u style="single">Is a vector corresponding to an estimate of said subset of symbols, and [X] is a matrix of spatiotemporal samples of bursts of said symbols.</u><u style="single">The receiver described in [23].</u><u style="single">[31]</u><u style="single">Decoding the burst of the filtered symbol comprises performing error correction on the burst of the filtered symbol.</u><u style="single">The receiver described in [23].</u><u style="single">[32]</u><u style="single">The receiver according to [23], wherein the interference suppressing means includes a single antenna interference canceling (SAIC) filter or a dual antenna interference canceling (DAIC) filter.</u><u style="single">[33]</u><u style="single">The interference suppression means is configured to use a higher order model than used to filter the burst of the symbol to filter the burst of the recoded symbol, [23]. Receiver.</u><u style="single">[34]</u><u style="single">A machine-readable medium with instructions for suppressing interference in wireless communication.</u><u style="single">The command is</u><u style="single">Receive a burst of symbols and</u><u style="single">A first multi-weight interference suppression filter is used to filter the burst of said symbol.</u><u style="single">Decoding the burst of the filtered symbol to generate data corresponding to the burst of the symbol.</u><u style="single">The data is encoded to generate a burst of recoded symbols.</u><u style="single">Based on the burst of the recoded symbol, a second plurality of weights were calculated for the interference suppression filter.</u><u style="single">The interference suppression filter with the second plurality of weights is used to filter the burst of the recoded symbol and then.</u><u style="single">Decoding the burst of the filtered recoded symbol,</u><u style="single">With code for</u><u style="single">Machine-readable medium.</u><u style="single">[35]</u><u style="single">The command is</u><u style="single">Generate multiple timing hypotheses for the burst of the symbol;</u><u style="single">Based on a subset of the symbol bursts, corresponding multiple weights for the interference suppression filter were calculated for each timing hypothesis;</u><u style="single">For each timing hypothesis, the interference suppression filter with the corresponding plurality of weights is used to filter the subset of the burst of the symbol;</u><u style="single">One of the plurality of timing hypotheses corresponding to the selection criteria is selected, and the selected timing hypothesis corresponds to the first plurality of weights;</u><u style="single">The interference suppression filter having the first plurality of weights is used to filter the burst of the symbol;</u><u style="single">With more code for</u><u style="single">The machine-readable medium described in [34].</u><u style="single">[36]</u><u style="single">The machine-readable medium according to [35], wherein the selection criterion is a mid-amble estimation error.</u><u style="single">[37]</u><u style="single">The mid amble estimation error is</u><u style="single">Determining the estimation channel corresponding to the timing hypothesis,</u><u style="single">Performing single antenna interference cancellation for said estimated channel to obtain an estimated midamble sequence, and</u><u style="single">Comparing the estimated midamble sequence with a previously known midamble sequence to determine the midamble estimation error,</u><u style="single">Is calculated for each timing hypothesis,</u><u style="single">The machine-readable medium described in [36].</u><u style="single">[38]</u><u style="single">The determination of the estimated channel corresponding to each symbol comprises selecting a predetermined number of adjacent symbols in the subset corresponding to the number of taps in the estimated channel.</u><u style="single">The machine-readable medium described in [37].</u><u style="single">[39]</u><u style="single">The generation of the plurality of timing hypotheses estimates the position of the first midamble symbol in the burst of the symbol and selects the subset of the burst of the symbol from the symbols centered on the estimated position. A machine-readable medium as described in [35].</u><u style="single">[40]</u><u style="single">For each timing hypothesis, the first plurality of weights</u><maths num="33"><img file="JP5512805B2_D0033.tif" /></maths><u style="single">Calculated by solving</u><maths num="34"><img file="JP5512805B2_D0034.tif" /></maths><u style="single">Is a vector corresponding to an estimate of said subset of symbols, and [X] is a matrix of spatiotemporal samples of bursts of said symbols.</u><u style="single">The machine-readable medium described in [35].</u><u style="single">[41]</u><u style="single">The second plurality of weights</u><maths num="35"><img file="JP5512805B2_D0035.tif" /></maths><u style="single">Calculated by solving</u><maths num="36"><img file="JP5512805B2_D0036.tif" /></maths><u style="single">Is a vector corresponding to an estimate of said subset of symbols, and [X] is a matrix of spatiotemporal samples of bursts of said symbols.</u><u style="single">The machine-readable medium described in [34].</u><u style="single">[42]</u><u style="single">34. The machine-readable medium according to [34], wherein decoding the burst of the filtered symbols includes performing error correction on the burst of the filtered symbols.</u><u style="single">[43]</u><u style="single">The machine-readable medium according to [34], wherein the interference suppression filter is a single antenna interference canceling (SAIC) filter or a dual antenna interference canceling (DAIC) filter.</u><u style="single">[44]</u><u style="single">The interference suppression filter is configured to use a higher-order model than used to filter the burst of the symbol to filter the burst of the recoded symbol, [34]. Machine-readable medium.</u>