TWI351825B - Selectively disabling interference cancellation ba - Google Patents

Description

IX. Description of the invention: [Technical field to which the invention pertains] 'The present invention and the bee; ^Wireless communication I;, more specifically, the present invention relates to a colored material on the side of the received data processed by the scale terminal in the wireless communication system Elimination of interference, etc. [Prior Art] Cellular wireless railroads support the world's most documented and unrecognized services. Although Lu cellular wireless ifH is accommodating to provide voice service, the need to hide and support the data, and the demand for the service is greatly developed with the increasing acceptance of the public and the widespread use of the Internet. The previous data communication services were provided through the wired network, and today's bee-gathered households require him to contact (4) communications. Many wireless users want to use off-the-shelf phones, personal wireless data assistants, wireless laptops and/or other wireless devices, and other data communication activities. The data communication needs for wireless communication systems are increasing with __ shift. _ Therefore, the financial communication secret of the company is in the process of continuous improvement and development in accordance with the difficulty of secret communication (4) communication needs. A cellular wireless network includes a "network infrastructure" that wirelessly communicates with wireless terminals within its corresponding service coverage area. The fine-path infrastructure generally includes the base stations distributed throughout the service area. Each base station supports wireless communication within the respective cellular unit (S-group magnetic zone). The base station is connected to base station control Hs (BSCs), and each base station controller controls a plurality of base stations. Each base station controls the read-to-axis intersection c (MSC) connection. Each base station controller is also typically connected directly or indirectly to the Internet. At 1351825, each base station communicates with multiple wireless terminals within its cell/magnetic coverage. #. The BSC is connected to the base station and routes voice communications between the MSC and the serving base station. Msc • Routing voice communication to another MSC or routing to ps.Bsc in the service base station and packet shouting connection due to the communication of the _ age group network, the base station to the wireless terminal transmission crane "forward link, , secret no miscellaneous basis (four) miscellaneous called the link. The wireless link between the base station and its wireless terminals typically operates in accordance with one or more operational standards. These operations are fresh in the way of the distribution, establishment, service, and disconnection of wireless ships. At present, the industry generally makes the (four) cellular standard a global mobile communication system (GSM). The GSM standard, or GSM for short, is very popular in Europe and even worldwide. B GSM is initially / applied to voice communication, and it can also support data transmission after improvement. GSM General Packet Radio Service (GPRS) service and GMS (or global) Enhanced Data Rate Forward (EDGE) service coexist with GSM, sharing GSM standard channel bandwidth, Slot structure and time slice timing ). The GpRS standards and standards can also be used as enhancement paths for other standards (mi(4)i〇n_), such as the Heart and Pacific Digital Cellular Standard (pDC). In order to improve the data rate of EDGE in the 200KHZ GSM channel, in addition to the standard Gaussian Minimum Shift Keying (GMSK) modulation of gsm, more advanced modulation, such as 8-PSK (8-ary phase shift keying) is used. . EDGE accepts nine spatial interfaces that are not (spontaneously and quickly selected), also known as modulation and coding schemes (MCSs), with varying degrees of error control. Na's transient needs, high MCS material (five) cs Μ) use 8-PSK (office data rate) modulation for air transmission, low MCS mode (mcs(4) 1351825 month/day correction replacement page using GMSK (low data rate) modulation for aerial Transmission - When the cellular phone is in receive mode 'The same channel and adjacent channels • The GMSK/8PSK signal appears as colored noise, even for the bee-off phone receiver in receive mode. The ability to better receive information sent to the cellular telephone, the cellular telephone should try to eliminate these interference signals. Existing techniques for eliminating these interferences include channel equalization of the received symbols. However, existing channel equalization techniques cannot effectively eliminate The noise of the same signal and adjacent channels. Therefore, the interference cancellation technology needs to be improved. SUMMARY OF THE INVENTION According to another aspect of the present invention, a multi-branch equalizer processing module is provided to selectively activate Interference cancellation to process received RF bursts, including: a channel length estimation module, configured to: determine a phase with the RF pulse The length of the channel is closed; when the length of the channel is less than the value of the predetermined channel, the channel is miscellaneously corrected, and the selected area is not processed by the drying process; 'Based on the known debugging phase, the opposite pulse train is equalized, and the data bits are extracted from the RF pulse train; and the first dan ♦ 胄 bl § in the channel length associated with the riding ship _于 所 , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , The wire is debugged according to the cut 1825 re-encoded m-bit taken from the RF pulse burst. Eliminate the colored noise threshold phase of the channel length axis group selectively initiating interference in the case of the town =:: : phase _ colored __ with a predetermined threshold value = = the face material _ 峨 编 编 compared to the above > Preferably, the predetermined channel length value is obtained based on the hilly channel. n tearing __ Saki-burst-related channel length adjusts the equalization of the first-equivalent n Preferably, the multi-branch equalizer processing module further includes: a deinterleaver that deinterleaves the data bits extracted from the first equalizer, and the channel decoder' Decoding the extracted data bits for decoding; performing re-encoding; re-encoding the re-encoding to generate re-encoded data bits;, interleaver, re-encoding bit and the known debugging sequence Row interleaving, generating a re-encoded burst, wherein the second equalizer may further deinterleave the another data bit based on the re-encoding pulse_the de-parent; row decoding the channel decoding Preferably, the H pair includes at least a portion of the other material level (four)-amplitude, and the channel length estimation module detects to continuously renew the received RF burst when the interference cancellation process is selectively initiated. Including the first group of radio frequency bursts. 9 1351825 The radio frequency pulse train received by the first equalizer and the first-fourth-fourth quadrant worm includes a first-, a minus-fire, and the t-the first- Group injection before the bit frequency pulse train; The burst and the burst group includes four radio frequency pulses _; or two bursts of half rate adaptive money rate encoded speech. According to an aspect of the present invention, a wireless terminal is provided, including: a radio frequency front end for receiving a radio frequency burst; a baseband processor, the front end of the communication front end is configured to generate a baseband signal from the radio frequency pulse train ▲ processor and RF channel length estimation module 'is used to: determine the channel length associated with the RF burst; when the channel has a large dispersion, selectively disable the interference age processing; weight multi-branch, etc. The processing module is connected to the baseband: the multi-branch equalizer processing module is connected to the baseband processor: the multi-branch equalizer processing module is further configured to include: 1 and - correcting H 'based on the known paste phase touch, riding the ship 4 D'Equal' and extracting the data bit from the RF burst; entering the second storage, based on including the known gift phase and re-encoded data bits = Γ Γ 进行 进行 </ RTI> wherein the at least partially re-encoded burst α process is de-generated, the second equalizer pairs the axis pulse into 1351825 and extracts another from the RF burst Data bit Wherein the baseband processor and the multi-branch equalizer processing module are combined to: generate a data block from a soft decision; or when compared to the radio frequency burst phase channel length and the predetermined channel length threshold When the _ is greater than the _, the data block is generated from the other soft decision. Deinterleaver 'deinterleaving the data block; channel decoding II, decoding (4) the material frame from the data block. The data block; re-encoder, re-encoding the resource to generate at least part of the The coded interleaver, the system is re-encoded (4) when the advance is wrong, the part is re-encoded. (10) When the result is determined to be 消除hs threshold threshold 2, the measured value of the colored noise associated with the radio frequency pulse burst is greater than the threshold value; The RF pulse train associated signal to noise ratio is less than the preferred 'the predetermined channel length threshold' based on the hilly channel compared to the financial 嗓 threshold. Preferably said channel length estimation module further adjusts the equalizer value of said first-equalization n based on a k-channel length associated with said radio frequency burst. Preferably, the continuation of the material feed _determination (4) the received __ string of the interference cancellation process includes the 11-frequency burst before the second group of shots and the burst group includes: 4 radio frequency pulses String; or 2 bursts of half rate adaptive multirate encoded speech. Preferably, the channel length estimation module further reduces the equalizer state of the first equalizer when the age is selectively activated based on the pass degree associated with the radio frequency pulse train. Preferably, said _burst comprises a parameter of a Gaussian minimum frequency shift keying symbol carrying said data and a gamma/Gaussian minimum fresh keying symbol. According to an aspect of the invention, a method for equalizing a received radio frequency burst includes the steps of: detecting a received radio frequency burst to determine a channel length associated with the radio burst; establishing a pre-channel length a threshold for selectively initiating an interference chirping process; determining a channel length associated with the radio frequency burst; comparing the channel length associated with the radio frequency burst with the predetermined channel length and said RF pulse Initiating interference cancellation processing when the string-related channel length is greater than the predetermined channel length threshold; debugging the first equalizer using a known debug sequence; using the secondizer to the received RF The pulse train is equalized; 12 1351825 performs the ^1~~1 on the radio frequency pulse train; and the plurality of radio frequency pulses Φ are decoded to obtain the extracted data bit element; from the extracted data bit Decoding an asset frame; wherein the 'interference cancellation process comprises: re-blocking the data t, to generate a re-encoded data bit; The material bit is interleaved; the second equalizer is debugged using the known debug sequence and the re-encoded data bit. • 1 (2), the second, the second, the fourth, the fourth, the __, the _, the _ Another data bit; deinterleaving the extracted another data bit; and extracting another capital from another data bit after the deinterlacing. Preferably, the method further comprises: determining a measured value of the colored arm tone with the radio frequency pulse; determining a channel ratio associated with the radio frequency pulse train. Selectively initiating the interference cancellation measurement value when the predetermined colored noise threshold is greater than the threshold value associated with the radio frequency pulse string; the predetermined signal to noise ratio threshold is less than the The RF signal is related to the signal to noise ratio and the threshold value. 13 1 Preferably the method further comprises: processing: Preferably, the method further comprises: a force to selectively initiate the side division, the frequency reduction pulse phase _ channel length . DESCRIPTION OF THE PREFERRED EMBODIMENT The present invention will be described with reference to the accompanying drawings. • The Nans Minimum Shift Keying (GMSK) modulation system can be used as a single-input dual-out system model in the real domain. The model (4) is virtual single-shot and dual-received. As described in the present invention, the multi-antenna interference cancellation technology is applied (4), and the system can fully solve the above requirements and other requirements. The equalizer processing module of the distribution day of the distribution day is used to eliminate the interference associated with the received RF burst. The multi-branch equalizer processing set includes a multi-processor processing branch. The squamous-reduction branch performs the mining of the Linlin Wei by the known modulation sequence. The knot obtained by the weaving is processed in a step-by-step manner, and is turned over to the second storage branch. The New Second Equalizer processes 77 outputs that equalize the received Langlai pulse φ to produce an aged interference signal, thereby improving the processing of the RF burst. 1 is a schematic diagram of a portion of a cellular radio communication system 100 that supports a plurality of wireless terminals in accordance with an embodiment of the present invention. The cellular radio communication system 100 includes a mobile switching medium (MSV) 101, a GPRS service support node/edge service support node (SGSN/SESN) 1, a base station controller (BSCs) 152 and 154, and base stations 1〇3, 104. , 105 and 1〇6. The SGSN/SESN 102 is connected to the Internet via the GPRS Gateway Support Node (GGSN) 14 on the 351825 112. The conventional voice terminal (2) is connected to the psTN (10). A Voice over Internet Protocol (VoIP) terminal 123 and a personal computer 125 are connected to the Internet. The MSC 101 is connected to a public switched telephone network (pSTN) 11A. Each base station 1G3·106 touch-scale cell group magnetic zone provides wireless communication service support. The wireless link, including the forward link portion and the reverse link portion, provides support for the wireless communication between the base station and the wireless terminals it serves. The same channel or adjacent channel signals generated by these wireless links will appear as colored or white noise. As mentioned earlier, these noises interfere with the desired signal. Accordingly, the present invention provides a technique for eliminating such interference in a low signal to noise ratio (SNR) or low signal to interference ratio (SIR) environment. The above wireless link can support digital data transmission, ν〇ΙΡ transmission and other digital multimedia transmission. The cellular wireless communication system 100 is also backward compatible to support analog operations. The cellular wireless communication system 100 also supports the GSM standard and the EDGE extension standard. The cellular radio system 100 also supports GPRS. Further, the present invention is also applicable to other standards such as a TDMA standard, a CDMA standard, and the like. In general, the present invention is applied to the digital communication technology to handle the identification and elimination of interference communications. The wireless terminals 116, 118, 120, 122, 124, 126, 128, and 130 are coupled to the cellular wireless communication system 10B via a wireless link with the base stations 103-106. As shown, the wireless terminals include cellular telephones 116 and 118, laptops 120 and 122, desktop computers 124 and 126, and data terminals 128 and 130. Moreover, cellular wireless communication system 100 also supports data transmission with other wireless terminals. As is well known, laptops 12 and 122, desktop computers 124 and 126, data terminals 128 and 13 and cell phones 116 and 118 are capable of "surfing", transmitting and receiving data communications, such as electricity, on the Internet 114. 1351825 Γ Γ 文 魏 魏 魏 魏 魏 魏 。 。 。 。. Most of the __ _ download speed is strictly strict, € slightly #继_. The county property's departure == supports the earning, and the terminal 116 loss can also support ‘. FIG. 2 is a functional block diagram of the wireless terminal 200. The wireless terminal shown in Fig. 2 includes an RF transceiver 202, a digital processing unit 2〇4, and various other components housed within the housing. The digital processing unit 2〇4 includes two main functional components, one is a physical layer processing, a speech codec, and a baseband codec module. The other is a co-processing and human-machine interface module 208. The digital signal processor is a physical layer processing, speech codec, and baseband codec age 206 t resident component, such as the _ instruction set computing (RISC) processor is a auxiliary operator, human interface module 2〇 The main components of 8. Also known as the no-line interface processor (RIP), the RKC processor is also known as the system processor. However, these names are not a limitation of the functionality of these components. The RF transceiver 202 is coupled to an antenna 203, a digital processing component 204, and a battery 224 that powers all of the components that are invoked by the wireless terminal 200. The physical layer processing, the speech codec, and the baseband codec module 206 are coupled to the coprocessing, the human interface module 208, the microphone 226, and the remaining 228. The co-processing, human-machine interface module 2〇8 is connected to various components, such as but not limited to a personal computing/data terminal device interface 21, a keypad 212, a user identification module (SIM) 213, a camera 214, and a flash memory. Body 216, SRAM 218, LCD 220, and LEDs 222. These components can support both static and/or dynamic images when the camera 214 and LCD 220 are connected. Thus, the wireless terminal 200 shown in Figure 2 can support audio and video services over a cellular network. 16 GS GS]V [The structure of the frame and the way in which the data block is transmitted through the gsm frame. The dirty matter with a duration of 2 Gms is divided into 4 sub-frames, each of which has a period of time O. The duration of each time slice is approximately 625us 'including the left part (10) Na), the right part (from the fine side) and the middle part (this). The left and right parts of the RF burst of the time slice carry data, and The middle part is the debug sequence. The GSM duck's 4 time slice RF bursts can transmit the - segment of the RLC block, a complete RLC block or two coffee blocks, depending on the side complement and scale, 弋 &quot; Through the first quarter of the duck time slice 0, the second time of the middle of the four knives, the third quarter of the time and the fourth quarter of the money Time slice G is transmitted. The data block A can carry one, one or two RLC blocks of the coffee block. In the same way, the data block B passes the i-th quarter-(four) time slice, the time slice in the first four-knife block, the third quarter shot of the time slice 1 and the fourth quarter shot. Time slice i is transmitted. Each set of time slices (ie, the G· shot each sub-time slice MCS mode is stable for the g_, but for the _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ group time chip MCS The pattern 'for example, the time slice of each sub-building is not the same as any of the time slices π of each sub-building. The rainbow block can carry voice data or other data. Figure 4 shows the mapping of data to radio frequency pulses. Schematic diagram of the process of the string. The original data is unencoded and has a data block header, which is used to perform external coding on the (4) block and recognize the error check of the data block. The outer rim operation generally makes the _ redundancy check ((10) Or Fire Code (FireC〇de). As shown in the figure, the external encoding operation appends the tail bit and/or block coding sequence (BCS) after the data 17 丄妁 1825. In the cs_i coding scheme, block coding and The convolutional coding encodes the header and the data together. In the non-cs_i coding scheme, the header and the data are generally encoded separately. The Falun error correction and error detection are a shortened binary loop code in the header. And add redundancy after each bit of the data The full error detection capability of the Farr code makes the undetected rate only possible. The block coding operation pairs (4) adding redundant bits for error checking, and calculating additional redundancy when making errors. Transmission error caused by the wireless channel. _ Error correction or coding scheme is based on convolutional coding. Convolution, some bits generated by the flat coder can be shrunk before transmission. Shrink processing can increase convolution The coding rate reduces the redundancy of each data block sent. In addition, the shrinking process can also reduce the bandwidth requirement, so that the convolutional coded signal can be made to conform to the available channel bit turbulence. The convolutional coded and contracted bits It is transmitted to the interleaver, which shuffles the different bit streams and divides the interleaved bit rate into four bursts as shown in (4). Art® 5 does not test Na Naifu The thorns are generally received. Four shed bursts are received and processed to form a data block. After receiving * RF bursts, σ is combined to form an encoded data block. Then the encoded data block is solved. Shrink processing (if needed), decoding according to the internal decoding scheme, and then decoding according to the external decoding scheme. The decoded (four) block includes the data block header and data. According to the data and the newspaper-encoding method, the partial decoding can identify the data.

6 is a schematic diagram of a process of restoring data from a transmitted voice. This is similar to the process of stealing. - Send the duration I is the voice (4) of the I 18 . In the burst, the 2° ms is the second half of the speech. For example, (4), - Group Na_ from the first language ': On _ 1Gms (milliseconds ), the first and second of the syllables and the second half of the 〇 = 〇 are half-coded and interleaved in the _ group 4 (four) pulse =. When the @KP burst is processed, the generated marshalling block generates a data arguing &quot;咐二半收音^ The first generated speech &amp; the first half can be combined with the second half of the speech t to generate and validate The data related to the arpeggio frame η. Less: ===phase_# material to re-encode to the voice_second half merge, + and financial touch burst recovery Use the loop m to generate a frame of data. The speech frame is validated and the voice is taken to generate an effective corpus. However, the shot is reconstructed for the valid one: = Dan is = the second half of the re-edited mother's tone can be used to generate the H handle °. The second half of the S frame is divided and interleaved, and the coded job string. Because of the linguistic exhaustion: knowing the tuning sequence, you can better debug the second equalizer processing branch. The string "wlAsZlT!T200 ^aAm ^ 对应 corresponds to the operation shown in Figure 8β (four) μ--corresponding time slice 19 1351825 two: gamma words. These operations (four) front-end, baseband processor and equalizer processing, And these operations are generally performed by one of the above components. However, this also • different components _ processing responsibilities may be different without departing from the scope of the invention. 2, as shown in Figure 8A, first, the rf front end Receiving a _string in the corresponding time slice (the step is called then (4) the front end converts the burst into a baseband. Step 8 〇 4). In order to complete the present invention, the front end sends an interrupt to the baseband. Step _). Therefore, as shown in FIG. 8A, the 拙 front end performs step 8 café 6. Then the 'baseband processor receives the baseband signal (step merging operation: front end: baseband processor or modulator/demodulator) Sampling the analogy with the base money to the dragon base greedy: the digitization of the input. After receiving the baseband signal (digital form), the baseband processor is obscured by the servant ^ _ 8ig). _ format blind check ^ _ _ money _ ageing. _ gsm fresh - a specific example in the 'modulation The format may be a Gaussian Minimum Shift Keying ((10)) modulation or a Phase Shift Keying (8PSK) modulation format. The baseband processor makes a decision (steps and continues to perform one of the two branches based on the detected modulation format). For the GMSK modulation format, the baseband processing baseband signal is inverse rotated and frequency scaled (step 814). Then, the baseband processing baseband signal performs the burst power = step_). Then, the page break connector A is connected to the map. In the case of the miscellaneous materials, the limit, the money, and the material are measured. Then, the calculation of the baseband line automatic gain control loop (step is called the domain, the baseband processor does determine the soft decision scale factor of the signal (step-like). After step 824, the baseband processor performs matching filtering on the baseband signal (step 826). 20 / 4 _·826 is hereinafter referred to as equalization pre-processing operation. The baseband processor performs the processing by these pre-equalization operations. The baseband signal. After the completion of the pre-equalization processing operation, the baseband processor issues a command to the equalizer module, and the work is performed by the equator. 2 will be introduced later in conjunction with Figure 9. Next, after receiving the command, the equalizer module performs a uniformization of the processed baseband signal based on different tunings such as 〇 modulation modulation and modulation. The equalization group receives the processed baseband signal from the baseband processor, sets or touches and performs maximum likelihood sequence estimation (MLSE) on the left portion of the baseband domain (steps are as shown in FIG. 3 above). 'The perna pulse train includes the middle part and the right part of the side portion of the data. In general, in the step micro, the equalization = 'equalizer module equalizes the right part of the processed baseband signal (step The equalization of the string corresponds to the butterfly part (four) butterfly. The pulse stringing _ 6 knows the sequence to proceed. However, the needle embodiment of the present invention can use re-encoded or partially re-encoded data to improve the equalization over===== process, wherein the first branch performs the equalization of the burst and the second = the first branch pair-group The processing result of the burst performs a second equalization process. The pulse train (four) Mei module sends an interrupt signal to the baseband processor, indicating that the soft-decision equalization 11 module receiving side, and the second data received from the equalizer module are soft left-base two: = The average phase (step 832). The soft decision of the smoke ^^, and other domain groups is subjected to frequency estimation and follow-up (steps). Step 21 1351825 - The trip or the operation of steps 854 and 836 is also referred to as "equalization post-processing". After the operation of the step is performed, the processing of the RF burst ends. _ 8A, when the blind detection is a bribe modulation in step _, the baseband processing awaiting module performs the flow on the right side of step 8U. First, the baseband processor performs a reverse _ and _ correction on the baseband (4) (step 818). The wire processor then performs a burst power estimate on the band signal (step 9). Then through the corresponding flow in the page change connector B, the baseband processor executes the time limit, channel, noise, and signal-to-noise ratio (step 84〇). Then in step 842, the baseband processor performs the calculation of the agc loop and then the baseband processor computes the decision feedback equalizer (DeciskmFeedbackEqualizer, ...) coefficients to be used by the equalizer module (step 844). The generation of this correlation process will be detailed later in conjunction with Figure 9. Next, the base processor performs an equalization operation on the base 2 signal (step 846). Finally, the baseband processor determines a soft decision scale factor for the baseband signal (step 848). Step 8 of the baseband processor 3 is performed on the 8PSK "Equilibrium Secretion, Operation. After the end of step 848, the baseband processor issues a command to the equalizer module to the processed baseband. The signal is equalized. After receiving the command from the baseband processor, the equalizer module receives the processed baseband signal, settings, and parameters from the baseband processor and begins equalization of the baseband signal. First, the equalizer mode The group cancer value is used in the equalization of the 8PSK modulated baseband signal (step 850). In the embodiment shown in the figure, the equalizer module is equalized using the maximum posterior probability (MAP). Then, the equalizer module equalizes the left and right sides of the processed baseband signal using the _ equalizer to generate equalization of the processed base 22 λ Ο Δ Ο λ Ο Δ = == _). Step_Complete·etc. To baseband. Issue interrupt field' indicates that it has completed the processing of the baseband signal after the processing. Then 'base_|| from conversion/receive soft^ to the soft outline峨_细触蝶靖2 _Γ)854)°·Riding 4 Qingqing^fine__°Tracking (step 736&gt; The operations of steps 854 and 836 are also referred to as equalization post-processing operations. After step 836, the operation of the particular RF burst is terminated. Although the above operations are performed by specific components without secrets, 'these operations are still This may be done by a non-component. For example, in some other embodiments, the equalization interpolation may be performed by a baseband processor or a system processor. Further, the decoding operation page may be executed by the baseband processing system. There are two methods for constructing a multi-branch equalizer processing module 900 for performing single antenna interference cancellation (SAIC) according to an embodiment of the present invention: 1) joint detection (JD); 2) blind interference cancellation ( BIC). The invention _ BIC method. The various parts shown in Fig. 9 may be hardware, executed by a processor (such as the shift or 2 〇 8 in Fig. 2) or a hard axis soft _ combination. The old processing module 900 includes a first equalizer processing branch and a second equalizer processing branch. The reverse rotation module receives the baseband burst hat (1) and orthogonal (9) age, and the baseband burst corresponds to FIG. 3-7. The pulse train described. To the rotating module, the received and Q pulse samples are reversed, and the Ϊ and Q pulse _ county (pulse) are generated. The channel length estimation module 906 ships some veins _ into the hiding _ determine the observation of the Wei Qiandao delay expansion 23 1351825 - Extensibility. Channel length or channel delay extension will affect the performance, power and complexity of the GSM/GPRS/EDGE receiver. The longer the channel length, the more channels need to be sliced in the first-equalizer_ (channeUip) estimates. This shows that in the case of a female-length thief sequence, the longer the channel estimation accuracy is, the smaller the propagation model is. The longest propagation model is the hilly terrain (HT) channel in the Gs hall dge specification. . In addition, the single-antenna interference cancellation () technique is difficult to apply to the HT channel because the number of equalizer coefficients of the HT channel is too large, and the coefficient of the coefficient is not very good. Moreover, in order to accommodate the long channel slicing, the complexity of the SAIC under the HT channel is relatively increased. Therefore, multi-branch equalizers need to be able to recognize channels with long delay spreads (such as HT channels) to perform SAic. When such a channel is identified, the S-branch equalization II can interfere with the ASIC and use only existing processing techniques. While existing processing techniques are only used for channels with long delay spreads, existing processing techniques such as Ητ channel' GSM/Gm/EDGE can benefit from known channel dispersion. For example, the MLSE receiver in the first equalizer processing branch 902 can adaptively select the number of green states for different channels, and the value of the maximum state of the ground for the use of the channel. This self-ship selection method improves the performance of small dispersion channels. In one embodiment, to accommodate the ΗΤ channel, a window with 6 channel slices is used to estimate the channel. The correct channel should be located where the window contains the maximum channel energy. The scale limit can be changed by searching for the time limit of the channel with smaller dispersion, such as in rural areas (1^) and typical urban areas (TU). These will be described later in connection with FIG. 10 and FIG. Figure 1 is a schematic diagram of the channel estimation window position of the small dispersion channel, such as ^ and 11; 24 pass [modulo 1_the length of these channels is smaller than the size of the window. If noise is not considered, the channel can be estimated to have multiple optimal window positions as long as the window towel includes 3 main channel slices. Figure 20 shows no two optimal positions. However, if the maximum channel energy standard is used to k _position' because there is noise, the window positions of the different bursts will be different. Figure 11 is a schematic illustration of the channel estimation window position for a large dispersion channel such as an HT channel. The channel is the same length as the window, so only the best window position contains all the channel energy. Therefore, the channel estimation time variation of the HT channel is small. • Based on the second-order statistics of the channel estimation time period, the dispersion of the channel can be determined. The following sections describe how to estimate time limits and generate statistics. The receiver estimates the execution time estimate for the channel of each burst. This time limit estimation operation has 3 steps. First, a cross-correlation is performed on a pulse train received near the intermediate position (touch (four) test) using a known debug sequence, as follows: .诘〆63 + female + /&gt;(/) ^ = • where y(D is Received GSM/EDGE burst, a1 is a known GSM/EDGE debug sequence with 16 symbols, and the result of hybridization is related. Next, 'the calculation of the channel is not long. The channel energy of staring. 4) Let (4) ".., 7 where e(i) is the channel energy. In this case, the timing of channel estimation is the position with the largest channel energy.

Vct = argmpe(/) 25 1351825 where η〇! is the estimated timing for channel estimation. When the receiver is working, it is connected to the base station, and the hold-segment can be obtained for a long time with the receiver time and the vertical loop. (4) The mean square error of sum is as follows: s = hLT) 2 Further taking the exponential average of the mean square error obtained from the nth pulse train can be obtained:

Save -£ave'^-(X) + €-a where α is the exponential constant. The smaller the constant, the larger the average. #图12 is a schematic diagram showing the time-averaged variance of different propagation models in (4).G1. It can be seen from Figure 12 that the mean square error of the HT channel is much smaller than that of the small dispersion channel such as the RA or TU channel. Therefore, if the mean square error and the predetermined channel length threshold are compared, the Ητ channel or other similar channel with long delay spread can be easily identified from FIG. For example, the predetermined channel length value may be τ1ΐΓ_τ=3. If the calculated mean square time difference is smaller than ThreShHT, the 'view~, _ channel has a larger dispersion. Lu therefore should stop using interference cancellation processing (SAIC) to prevent receiver performance degradation. On the other hand, when eave&gt;ThreshHT, the number of states in the existing concealed receiver should be reduced to improve performance. Returning to Fig. 9, the switcher 9〇9 makes the interference cancellation process valid or invalid based on the decision. In order to determine whether to make the interference cancellation process (ie, the second equalizer processing branch) valid or invalid, these criteria can be logically combined with other factors, such as but not if the SNU is colored. The noise and the received bursts are determined by chewing tone limits or interference limits. &lt; One embodiment towel H processor processing branch 902 includes conventional pulse train equalization 26 1351825 . These samples will then be equalized, while the other samples constitute a packet, for example (4). Under certain operating conditions, in addition to the equalization of the burst level, the iterative processing of the second equalizer branch is performed. The pulse equalizer, including the WQ finite pulse filter 9〇8 and _ and the least squares estimation (MLSE) equalizer 912, processes each burst received from the inverse rotation module. Group 913 uses the known debug sequence (TS) of the intermediate components within each burst. On the other hand, these components can be debugged on multiple pulses. The first equalizer processing branch 〇2 generates a soft decision in which a plurality of soft decisions represent decoding _ each refusal element. Each soft sample is provided to the interleave, which is interleaved and sent to the channel decoder 916. The channel decoder 916 decodes the capital from the soft sample (i.e., a plurality of soft samples representing each data bit are decoded by the decoder to generate decoded hard bits). The data packet generated by channel decoder 916 is acknowledged by re-encoder 918 and re-encoded&apos; to produce re-encoded data bits. The interleaver 92() receives the re-encoded data bit ^' to generate a re-encoded data burst. The second equalizer processing branch 310 is then debugged using the recoded data burst and the known debug sequence. The second equalizer processing branch 904 includes register-like and I and Q FIR filters 924 and 926 for storing a plurality of bursts. The ϊ and Q filters are similar to the debug module 928 using a known debug sequence and at least partially re-encoding the pulse (four) line debug. By this method, the second equalizer processing branch uses the at least partially re-encoded data and the known debug sequence to perform the Ϊ and q RF filters. When the switcher_starts the second equalizer processing branch according to the round-out of the channel length estimation module 907, these operations improve the handoff of the bursts processed by the 222. ...Q chopper is used to process the stored bursts after debugging. The resulting results are summed in adder 93, which results in another set of soft samples, which are then provided to interleaver 914 and channel decoded $916 for processing to generate another set of data bits. Figure 13 illustrates the first branch of the multi-branch equalizer of Figure 9 in more detail. In the case of ideal tuning, the 2-branch linear equalizer and the decision feedback equalizer (dfe) can achieve __ performance improvement over conventional receivers. Since there are only a few adjustments, as shown in the figure, the first-processing branch allows the thief to adjust the _-surface waver _ to debug the feedforward choppers 908 and 910. Figure 14 illustrates the second branch of the multi-branch equalizer of Figure 9 in more detail. After channel decoding, the data is re-encoded and used to debug the 7-slice linear equalizer (10) and you. The reason why the second branch equalizes n is to be used for _interlacing. The re-encoded bit associated with a speech block may only provide half of the burst (even the data bit). The decision feedback equalizer (10) E) requires continuous, continuous sampling for the feedback filter. Moreover, the linear equalizer (10) is selected to reflect the H(10)E). In the embodiment of the fine _ full re-editing element, the Pengke is selected for the second branch. The temporary storage (4) is located downstream of the switch 909. The register 922 can also be placed upstream of the cut shop 9〇9. In this way, regardless of the state of the pass count card pool 9G7 or the city (four) device 9, the received RF bursts are stored in the memory. In this way, when the second branch of the multi-branch equalizer is activated by the channel length estimation module 9〇7 or the output of the switching controller, the pulse train can be used immediately. Figure 15 is the use of interference cancellation. The process of equalizing the received bursts is 28 = Γ _ receives a plurality of bursts, and then in the step as before = = 2 turn. handsome, make the material - the first of the cans The equalization sequence pair: the first burst is processed, and in step 1506, the known debug: J== processing branch is used for debugging. The received burst, which can provide the branch inner branch and the first equalize 11 processing branch. Store the connection in the second equalizer processing memory or other memory location The RF pulse train is used to perform the sequence H meso + ', the device processing sub-divided has been based on the known debugging pulse train rrmr bribes after the balance of the heart, the commission in the cigarette cap it is interlaced two 1514: Γ 2 Decoding is performed to obtain the extracted data bit. Step 'Wei extracted A's material touch material decoding production, and the heart loses her handsome it. The current data obtained from the group is obtained from the previous _burst I', ~ and 'generate-a valid voice towel. New Zealand voice towel 贞 re-encoded to entangle the money gamma in the dump (10) _ find processing, the birth department == ΙΓ butterfly right (four) side of the 贞 include processing = two from the memory Zhao Zhong re-acquires the w burst and uses the second equalizer to use the second-order filament 1 to extend the steps including acquiring one or more bursts and rushing the second-order 彳= branch for processing. In step 1522, the _coded data The pulse is used to use the first object. Miscellaneous (four), in the 1524, the second equalizer processing branch can be used to equalize the good burst, wherein the second equalization 29 1351825 processing branch is not only based on the known debug sequence. Based on the initial output of the channel decoder, some partial re-editing The data bit is used to carry the thief. This makes the second wire processing eight = can not use the known debugging sequence and use the re-encoded data bit to better balance and debug, because the domain provides better than the first equalizer processing The branch is better rounded. The second equalizer processes the branch to generate another set of soft decisions. After the decompression of the interleaved step 1528 of step 1526, another data frame is generated in step 153. Step-by-step introduction to the minimum level material (LS.) method of closing the material 'This method is used in E. The money, _ Lai sequence estimates the channel. Then, when it is the front town of the brain The parameters of the tiling device and the MLSE are calculated when the anti-mineral wave device is used. The problem with this indirect debugging method is that the channel estimation (CE) performance is poor due to (4)^ = often working in a low arc, and the channel estimation error is poor. It is passed when calculating the chopper coefficient. The signal model of the MLSE input in Figure 13 is considered to be the lsi channel plus noise. Assume that the impulse response of the f-feedback filter is _増, bM}, and the purpose of the debugging is to obtain the pre-cord filter coefficient 〜~fl(Lr lion), and the number of the MLSE parameter b corresponding to the received signal. Based on the above mode, the noise of the MLSE wheel is derived from: /=〇 where X1 and X2 are the reverse rotation outputs J and Q, respectively, s is the debug number, and d is the system delay. Expressed in vector form: 30 1351825 n(k+l) η(]ς+Ν)_ x^k+d) x^k+d+ΐ) XjQc+d+N)

Xi(k+d-Lf+\) ..·工1(女+ Λ?+1 - Zy+1)Too 2(g+3 + 1) Xtik+d+N-^+^^lc+d+N )... ^(k+d-Lf+l) ^{k+d+l-Lf+X) x2(k+d+N-Lf+\) /ι(〇) AiLf -Ο λ(9) s(k) ϊ( Α:+1) ...s(]c-Lb+\) ... s(k+l-Lb +1) 'ms(k+N) -. s(k+N-Lb+l)f^Ll&gt; ^ For convenience, the lowercase letters in black are vector, and the uppercase letters are matrix: n=Xf-Sb The task of the equalizer is to find f and b that minimize the noise of the MLSE input, min |ju|| 2 Since the number of debug sequences is limited, the joint optimization of £ and b is sensitive to voice. The following describes the sub-optimal method for reducing the material parameters of the occupational wave device f. The cross-correlation between the pre-filter output (Xf) and the debug symbols can be performed on the coffee channel. In this case, b is indicated. In-service wave device, what is the source (four) material as_GE), coffee listening and listening b = S+Xf number most ==_4 (P one (four) can change can "the letter min f

l|Xf'SS+Xf||2 = min^ss+)Xf||2 = minrA 31 1351825 One of them is A=x, (1 - s s+) x, and 〇 is the transpose operation group (10)). • To avoid trivial solutions, you need to apply constraints to them. The two commonly used constraints are unit norm (Unit_n_) constraints and linear constraints. When the constraint norm is 】, the optimal solution is the eigenvector of A corresponding to the smallest eigenvalue: f = eigvec(A) Linear constraints can also be applied to f. For example, 'the last element of [) can be fixed to ^, that is, the i-th slice of the MLSE channel b is 丨. c is the earthy line of s+x. Thus, the line constraint is as follows: Thus, the optimal solution is:

The f=AV linear constraint is usually better than the unit norm constraint. In linear constraints, if the first slice is chosen to be one, the above-mentioned minimum standard is equivalent to the DFE standard. Diagonal fill is used in the high SIR range. 1 Under HT channel conditions' or other similar channels with long delay spread, interference cancellation processing can make performance worse than conventional processing. Long delay channels also produce large attenuation due to the shorter pre-filter length. To solve this problem, a converter is added to initiate or disable interference cancellation processing. The switch can be any combination of SNR, colored noise discriminator and channel profile detector. 16A and 16B are flow diagrams for using a channel estimation module to initiate or disable interference cancellation equalization processing to equalize received RF bursts. In step 16A, a plurality of bursts are received, and then in step 1602, the reverse rotation processing as described above is performed. Step 32 1351825 Help determines the estimated channel length or channel dispersion value, such as the mean square error described in the figure. Handle the secret towel, the system - #化_Figure (1), the processor processing branch handles the RF pulse train, the first equalizer processing branch is debugged in step = using the known debug sequence ship. The received pulse train can be provided to the first stage buckle and the second equalized H processing point A. The received egg burst is stored in the second #store processing branch using a scratchpad or other memory location. The rational branch performs the parent error and decoding on the smoke_, leg, and 16M_receive_burst. In the secret step, from the extraction _ level cap data ^ judgment point 1618, it is more advantageous to determine the estimated channel length or the measured value of the channel chromaticity. If the job is profitable, then only the first material = branch generation data is enough. However, if there is no advantage in the value of the measured value in step i6i8, the second equalizer handles the branch. When the signal is related to the GMSK letter (four) 妓 this case. m code data bit. In the speech building process, it is necessary to perform interleaving processing from the current group egg =: Γ voice one code generation re-encoding bit 2 to the new frame 7=24 to generate a re-encoded data burst. Application m The re-encoded data burst includes partially re-encoded bits. Shout 2^, re-acquire the W burst from the memory, and use the second equalization to rotate the (10) burst and use the second °" light processing. In step 1628, 'provide the recoded data burst 33 1351825 Debugging the branch of the second equalizer processing. In this way, it is stored in the memory. • The string can be processed in the step by using the second equalizer to process the branch. The processing branch is not only based on the known The sequence is also debugged based on less-partially re-encoded data bits, such as channel decoding, which enables the second equalization = branch to utilize both the known debug sequence and the re-encoded data bits to balance and debug the I. Therefore, it is possible to provide a better rounding out of the processing branch of the first equalizer processing branch, and to generate another-salt decision by the processing of the intersection of the first and the sixth (6) 4 in step 1632, and then generate the generated in step 163. In addition, the present invention provides a multi-branch equalizer processing module for receiving interference associated with a radio frequency burst. The multi-branch equalizer processing mode includes: Processor processing The equalization of the financial branch. The first-complex is debugged by the known debug sequence and the received financial burst is equalized. The soft sampled kernel of the soil is converted into: _ meta. The paste sample and the channel solution are performed. Processing, thereby generating a data bit from the soft sampled sample: a code block. The recoder re-encodes the solution to generate a re-encode or at least ^^' generate a at least partially re-encoded burst. The equalizer processing branch uses the at least partially re-encoded data bit to debug the linear equalizer therein. The received shrimp or the like is stored in the register, after debugging the linear equalizer, The == takes the received w burst from the register _ which performs 2 processing reads to generate another-soft sampled sample money t, and then converts to another 枓h. The other-soft sampled sample is described by The deinterleaver and channel decode 34 1351825 will generate another decoding, which can eliminate the processing precision of the interfering signal received by the interfering signal. The second equalizer processing branch can be Selecting the ground based on the measured channel conditions Invalidation is invalid. With long delay ship channel conditions ^ need to eliminate interference handling, _ this will make the performance worse than the transmission method. Those of ordinary skill in the art will recognize that the term "substantially, or "about," As can be seen here, the 'm-line terminology provides an acceptable tolerance. This industry-acceptable tolerance ranges from less than 1% to · and corresponds to, but is not limited to, component values, integrated circuit processing fluctuations. Temperature fluctuations, rise and fall times, and/or thermal noise. It will also be apparent to those skilled in the art that the term "operably connected, as may be used herein" includes passing through another component, component, circuit. Or module direct connection and indirect connection, where for the indirect connection, the intermediate insertion component, component, circuit or module does not change the information of (4) 'but its current level, voltage level and/or power level can be adjusted. As will be appreciated by those of ordinary skill in the art, the inferred connection (i.e., the component is inferred from the #7G piece) includes the same as the "operably connected," Indirect transmission, as will be appreciated by those skilled in the art, the term "ratio, as it may be used here, refers to two or more components, a project, and a signal 2 comparison to provide a desired relationship. For example, 'when the desired relationship is the vibrating towel 4' of the signal i with 2, when the amplitude of the money i is greater than the sum of the money 2 or the signal 2, and the field is less than the amplitude of the signal 1, a _ comparison result can be obtained. The present preferred embodiment has been described for the purposes of turning and description. Alum's exhaustion of the invention is based on the above, and it is also possible to modify or change the actual payment of the present invention based on the above-mentioned Le Rong Chen or according to the present invention. For the purposes of 35 1351825 - the explanation of the present invention is explained in the light of the principle and the actual turning, so that those skilled in the art can carry out the invention. It is apparent that the scope of the invention is defined by the claims and their equivalents. In addition, various modifications, changes and changes may be made to the present invention without departing from the scope and spirit of the invention. BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a schematic diagram of a portion of a cellular wireless communication system supporting multiple wireless terminals in accordance with the present invention; FIG. 2 is a functional block diagram of a wireless terminal constructed in accordance with the present invention; FIG. 3 is a GSM Schematic diagram of the structure of the Pa and the manner in which the data block is transmitted by the Na?? Figure 4 is a schematic diagram of the downlink transmission form; Figure 5 is a schematic diagram of the process of restoring the data block in the RF-series signal of the New Zealand series; Schematic diagram of the process of recovering speech data from the series of RF burst signals; Figure 71 Schematic diagram of the process of recovering bursts from data or speech_numbers; • Operation flow from the wireless terminal when receiving and processing the shots Figure 8 is a flow chart of the operation of the wireless terminal when receiving and processing the radio frequency burst; Figure 9 is a block diagram of the multi-branch pulse train equalization (4) according to an embodiment of the present invention; Estimate the position of the window (wind〇wp〇siti〇n) 围 · 围, Figure 11 示意图 large dispersion channel esoteric channel estimation window pulse position; Figure 12 is under different propagation models Schematic diagram of a mean square error; FIG. 13 is a schematic diagram of a pulse equalization (four) according to the present invention; 36 1351825 FIG. 14 is a schematic diagram of a pulse equalizer according to an embodiment of the present invention; An operational flow diagram of an embodiment; 'Fig. 16A' is an operational flow diagram in accordance with one embodiment of the present invention; and Fig. 16B is an operational flow diagram in accordance with an embodiment of the present invention. [Description of main component symbols] 1 〇〇 寓 无线 无线 无线 移动 移动 移动 移动 移动 移动 移动 移动 移动 移动 GPRS GPRS EDGE EDGE EDGE EDGE EDGE EDGE EDGE EDGE EDGE EDGE EDGE EDGE EDGE EDGE EDGE EDGE EDGE EDGE EDGE EDGE EDGE EDGE EDGE EDGE EDGE EDGE EDGE EDGE EDGE EDGE EDGE EDGE EDGE EDGE (Public Switched Telephone Network) 112 GPRS Gateway Support Node (GGSN) 114 Internet 116, 118 Cellular Mobile Phone 120, 122 Laptop 121 Voice Terminal 123 Voice (IP Voice) Terminal 125 Personal Computer 124, 126 Desktop Computer 128, 130 data terminal 200 wireless terminal 152, 154 base station controller (MSC) 202 radio frequency transceiver 204 digital processing component 208 human interface function block 212 keyboard 214 camera 218 static memory (SRAM) 222 light emitting diode (LED) 226 microphone 203 antenna 206 baseband encoder/decoder (CODEC) function block 210 PC/data terminal device interface 213 user identification card (SIM card) 埠 216 flash memory 220 liquid crystal display (LCD) 224 battery 37 1351825 900分支 branch equalizer processing module 904 second equalizer processing branch 907 channel length estimation module filter 909 switch 228 speaker 902 first equalizer processing branch 906 anti-rotation module 908, 9101 and q finite impulse response (ρπ) 907Α handover controller 912 Minimum Least Squares Estimation (MLSE) equalizer 914 deinterleaver 918 re-encoder 922 buffer 928 training module 913 training module Group 916 channel decoder 920 interleaver 924, 9261 and Q group pulse filter (FIR) 930 adder

38

Claims (1)

«51825 X. Patent Application Range · Selectively activate interference cancellation to process received 1. A multi-branch equalizer processing module RF burst, including: Channel length estimation module, for: determining and RF The length of the channel associated with the burst; $ has a larger Newlin, and the selection is invalid; The first equalizer 'equalizes the debug based on the known debug sequence, equalizes the shot string, and extracts the data bit from the RF burst +; equalizes 11, in the channel with the RF burst _ And when the length and the predetermined channel length threshold are less than the value, the at least one of the known ship sequences is equalized, and another data bit is extracted from the RF pulse train. yuan. 2. The multi-branch equalizer processing module of claim 2, wherein the Lu's second equalizer is further configurable according to a re-encoded data bit extracted from the RF burst. . 3. The multi-branch equalizer processing module according to claim 1, wherein the length estimation and recording of the interference are selectively initiated in the following cases; the shaving process: and the radio frequency pulse __ The colored noise _ the woven woven ridge has a colored sound threshold greater than the inter-value; the letter-arm ratio associated with the surface burst is less than the threshold compared to the predetermined signal-to-noise ratio. 39 1351825 _ 丨 日 修正 修正 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 多 多 多 多 多 多 多 多 多 多 多 多 多 多 多 多 多 多 多 多 多 多 多 多 多The multi-branch equalizer processing module of claim 1, wherein the channel length estimation module further adjusts the equalization of the first equalizer based on a channel length associated with the radio frequency burst The number of states. 6. A wireless terminal, comprising: a radio frequency front end for receiving a radio frequency burst; a ** processor with a radio frequency communication connection, the baseband processor and a radio frequency front end for generating from the radio frequency burst a baseband signal; a channel length estimation module, configured to: determine a length of a channel associated with the radio frequency burst; the length of the channel is greater than a predetermined channel length, that is, the channel has a larger When the level is selected, the dry selection process is selectively invalidated; a multi-branch equalizer processing module coupled to the baseband processor and the mode selection module, wherein the multi-branch equalizer processing module receives baseband transmission from the baseband processor, and the multi-branch equalization The handler handles the touch-step including ··° first equalization H, based on the _ trial phase, the job balance, and extracts the data bit from the RF burst; a second equalizer is based on Included by the known debug sequence and the re-encoded data bit to the > partially re-encoded burst, paste the slogan, and extract the other data bit in the private ship; 40 丄 351825 Wherein the baseband processor and the multi-branch equalizer processing module are combined for generating a data block from the soft decision; or » when the length is the same as the length of the channel A data block is generated from another soft decision when the predetermined channel length threshold is greater than the threshold. Deinterleaver, deinterleaving the data block; channel decoding ^, the ship's data block Lin's dragon ship line decoding; re-encoder, re-encoding the data frame to generate at least partially re-encoded data Block; the interleaver 'rides at least partially re-encoded blocks for interleaving to generate the at least partially re-encoded bursts. 7. The wireless terminal of claim 6, wherein the channel length estimation module selectively activates interference cancellation processing in the following cases: a measured value of the colored operational sound associated with the radio frequency burst When the threshold value is greater than the predetermined colored noise value, the signal-to-noise ratio associated with the radio frequency burst is less than the predetermined signal-to-noise ratio value. &gt;, a method for equalizing a received (four) frequency pulse train, comprising the following steps: · receiving a radio frequency pulse train to determine the length of the channel with the radio frequency burst _ channel length; Starting the shame processing; determining the length of the channel with the radio frequency burst _ channel length; comparing the length of the __ wheel length object 峨 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 When the predetermined channel length is said to be the inter-value value, that is, when the channel has a large dispersion degree, select "," "off interference cancellation processing, start interference cancellation processing; Debug with the known debug sequence - etc. The equalizer equalizes the received RF pulse train; deinterleaves the RF pulse train, and === rushes the string to obtain the extracted data bit; The towel erasing ^ data building; wherein, the interference cancellation processing comprises: ten kilograms of data touch re-encoding, generating re-encoding data bits · interleaving the re-wrapping touch; ϋ the known phase and the Re-encoding chaotic Debugging the second equalizer. The device performs - on the received radio frequency burst, and deinterleaves the extracted another data bit from the other data bit in the deinterleaved 9. If the patent application scope is 8', the bedding frame. In the above, the method further comprises: determining a measured value of the colored noise associated with the radio frequency burst; determining a channel ratio associated with the radio frequency burst. 10. The method of selectively initiating interference cancellation processing as described in claim 9 of the patent application, wherein the method further comprises: when the measured value of the phase-phase material has a greater than the threshold value compared to a predetermined colored value; Hungary 42 1351825 The signal to noise ratio associated with the radio frequency burst is less than the predetermined signal to noise ratio threshold. 43