Title of Invention

A METHOD FOR RECOVERING DATA AND A RECEIVER UNIT TO PROCESS A PHYSICAL CHANNEL

Abstract Techniques for recovering data transmitted on a physical channel in which channelization code is not known at the time of the data recovery. A modulated signal is received and processed to provide received samples (612). A hypothesized channelization code (e.g., an OVSF code in the W-CDMA system) is selected (616) and used to process the received samples to generate partially processed symbols (618). The hypothesized channelization code is a 'base' code that can be used to generate all possible channelization codes that may have been used for the physical channel. Intermediate results representative of the partially processed symbols are stored (622) and, upon determination of the actual channelization code (624), further processed (626) in accordance with the actual and hypothesized channelization codes to provide the final results. The additional processing includes partitioning the intermediate results into sets, scaling each intermediate result in a particular set with a scaling factor (+1 or 1) determined by the actual and hypothesized channelization codes, and combining the scaled results in each set to obtain a final result. In the STTD mode in the W-CDMA system, the final results from multiple actual OVSF code intervals can be selectively combine to obtain a recovered symbol (628).
Full Text

METHOD AND APPARATUS FOR PROCESSING A PHYSICAL CHANNEL WITH PARTIAL TRANSPORT FORMAT INFORMATION
BACKGROUND OF THE INVENTION
L Field of the Invention
The present invention relates to data commvmicatioii. More particularly, the present invention relates to a novel and improved method and apparatus for processing a physical channel "with partial transport format infonnation.
n. Description of the Related Art
A modem day commimications system is required to support a variety of applications. One such communications system is a code division multiple access (ODMA) system that supports voice and data commuxucation between users over a terrestrial link- The use of CDMA techniques in a multiple access communication system is disclosed in. U-S, Patent No. 4,901^7, entitled "SPPJEAD SPECTRUM MULTIPIJE ACCESS COMMUNICAITON SYSTEM USmC SATELLITE OR TERKESTKEAL REPEATERS," and U.S. Patent No. 5.10v3459, entilied "SYSTEM AND METHOD FOR GENERATING WAVEFORMS IN A CDMA CELLULAR TELEPHONE SYSTEM/ both assigjned to the assignee of liie present invention and incorporated herdba by refei'ence-
A CDMA system is typically designed to conform to one or more standards. One such first generation standard is &e •TLA/ElA/15-95 Remote Station-Base Station Compatibility Standard for Dual-Mode Wideband Spread Spectrum CeHular System," hereinafter referred to as the I&-95 standard and incorporated herein by reference. IS-95 compliant CDMA systems are able to tranfacrdt voice data and packet data. A newer generation standard is offered by a coiisortimn named "3"* Generation Partnership Project" (3GPP) and embodied in a i^ of documents including Document Nos. 3G TS 25^11, 3G TS 25:212, 3G TS 25-213, and 3G TS 25.214, wMch are readily available to the public The 3QPP standard is hereinafter refared to as the W-CDMA standard and incoiporated haein by reference.
The W-
"dedicated" channel on the dovnilixjk (Le,, from the base station to the losa:) and the uplink (i.e., from the liser to the base station) for the duration of the call. The dedicated channel can be used for voice commizixication or transmission of snutU amounts of padcet data, and linis has a (relatively) low bit rate. When 1±ie bas*2 station has a large amount of data to transmit, a "shared" diannel may be assi^ed and used for the data transmission. The shared channel is allocated and de-allocated to users as needed.
. The shared channd has a structure that supports a variety of uses and a bit :rate that can be varied between a range of values (e.g., from 15 kbps up to 1.92 Mbps)* The shared channel bit rate can also be dynamically changed from rad:.o frame to radio frame, with each radio frame being a unit of transmission cov'sring 15 time slots and each time slot corresponding to 0-667 msec.
In accordance with the W-CDMA standard, certain information necessary to properly recover the data transmitted on titie shared channel is pro-aded on the dedicated dhannel assigned to the user. For example, in certain instances, the bit rate of the radio frame and the channelization code (akin to titie Walsh code in the IS-95 CDMA system) used to channelize ttie shared diaimel are provided on the dedicated channel at approximately the same tibne as the data transmission on the shared channel- Consequently, the receiver unit is not able to process and recover the data on the shared channel in real-time.
Thus, techniques that can be used to efficiently process a physical chaimel, such as tine shared channel in the W-CDMA system, when some chaiacteristics of the physical channel are not Imovm are highly desirable,
SUM3VIARY OF THE INVENTION
The present invention provides techniques to process a physical channel in whidi at least some of the information needed to fully process the physical channel is not available at the time of processing. In accordance v/ith the invention, for a parameter that is unknown (e-g., a channelization code), a determination is xnade as to the possible values for that parameter. A ^ hypottiesised parameter value is titien selected and used to process the physical channel to provide intermediate results, which require less storage than the imp;roccssed samples. The hypothesized parameter value is selected such that the intermediate results can further be processed, when the actual parameter value is known, to obtain the final results.

Ceartain aspects of the invervtioii are especially suited for processing a received signal that has been processed and transmitted in accordance *with a spat^e time block coding transmit antenna diversity (STTD) mode defined by the W-CDMA standard. For some data transmissions in the W-CDMA system, the actual channelization code may not be knovsm at the time of the data reccwery at die receiver unit. For these data transmissions, a hypotheazed chajineHzation code can be used to partially process the received samples, to reduce tiie amount of data to be stored before the actual channelization code can be determined. The hypothesized channelization code is a fraction of, and can be used to generate the actual channelization code.
An aspect of the invention provides a method for recovering data transmitted on a physical dxaixnel in whidi a channelization code used for the physical channel is not knovsm at the time of the data recovery. In accordance vdfii the method, a modulated signal is received and processed to provide samples. A hypothesized channelization code is then selected and used to process the samples to generate partially processed symbols. Intermediate results representative of the partially processed symbols are stored and, upon determination of an actual channdization code used for the physical channel, fuxtlier processed in accordance with the actual arxd hypothesized channelization codes to provide the final resxilts.
The hypothesized channelization code is a 'base" code that can be used to generate all possible channelization codes that may have been tased to process the physical chaimel. The hypothesized channelization code has a length that is shorter or equal to that of the actual channelization code. For a W-CDMA system, the hypothesized diannelization code can be an orthogonal vari Various aspects, embodiments, and features of ihe invention are described in further deta£ below.
BRIEF DESCRIPTION OF IHE DRAWINGS
The features, nature, and advantages of the present invention will become more appar^xt from the detailed description set forth below when take!\ in conjunction with the drawings in which like reference characters identify correspoitdingly throughout and wh^ein:

FIG. 1 is a diagram of a spread ^ectrum communications system tiiat supports a nuiriber of users;
FIG* 2 is a sirr^lified block diagram of an embodiment of the signal processing for a dovmlink physical channel;
FIG. 3A is a block diagram of a modulator, a transmitter, and antennas that support transmission of a physical channel over two antennas;
FIG, 3B is a block diagram of an embodiment of a portion of the transmitter;
FIGS. 4A and 4B are diagrams that illustrate the encoding performed by an STTD encoder usm^ bit notation and complex notation, respectively;
FIG- 5 is a diagram that illustrates the generation of the OVSF codes;
FIG. 6 is a flow diagram of the processing of a modiJated (spread spectrum) signal in accordance with an embodiment of the invention;
HG. 7 is a block diagram of an embodiment of a portion of a receiver unit that can be i:ised to receive and demodulate a physical channel, including one transmitted feom multiple tranKnit antennas in the STTD mode of the W-CDMA standard;
FIG. 8 is a block diagram of an embodiment of a finger element that can be used to implement one of the finger elements in FIG. 7; and
FIG- 9 is a diagram that illustrates the processing of two symbols using an artual OVSF code and a hypothesized OVSF code.
DETAILED DESCRIPTION OF THE SPEOnC EMBODIMENTS
FIG* 1 is a diagram of a spread spectrum commimications system 100 ihat siq?ports a nimiber of users. System 100 provides communication for a number of cells 102a through 102g, with each cell 102 being serviced by a corresponding base station 104, Various remote stations 106 are diq)ersed throughout title system- In an embodiment, each remote station 106 com^nunicates with one or more base stations 104 on the downlink and uplink at aiiy given moment, depending on whether the remote station is in soft handoff^ The downlink refers to transmission from the base station to the , remote station, and the uplink refers to transmission from the remote station to the base station. The downlink and uplink respectively correspond to the forward link and reverse link in the IS-95 CDMA system. System 100 may be desi«|?ned to support one or more CDMA standards, such as the IS-95 standard, the W-CDMA standard, other standards, or a combination thereof-

As shown in HG. 1, base station 104a transmits data to remote stations 106a and IO63 on live downlink^ base station 104b transmits data to remote stations 106b and I06j, base station 104c transirdt& data to remote station 106c, and so on. In HG. 1, the solid line with the arrow indicates a transmission from the base station to tfoe remote station. A broken line x^ritil the arrow indicates that the remote station is receiving the pilot signal, but no ttser-spedfic data txaraanission, from the base station. The uplink communication is not shown in FIG. 1 for simplicity-
in certain transmission modes in the W-CDMA standard, a remote station can receive mtrltiple transmissions from multiple antennas of a single base station for certain physical channel types induding the (physical) downlink shared channel (DSCH)- As shown in FIG. 1, remote station 106a receives multiple trananissions from base station 104a, remote station 106d receives mtilliple transmissions from base station 104d^ and remote station 106f receives midtiple transmissions from base station 104f.
HG. 2 is a simplified block diagram of an embodiment of the signal processing fox a downlink physical channel. At the transmitter unit, data is sent, typically in packetS; from a data source 212 lo an encode 214. Encoder 214 performs a number of functions, depending on fhe particular CDlvIA system or standard being implemented- Sudi aicoding functioits typically include formatting each data packet with the necessary control fields, cydic redundancy check (CRC) bits, and code tail bits. Encoder 214 then encodes one or more formatted packets with a particular coding scheme and interleaves (or reorders) the symbols within Ihe aicoded packets. Encoder 214 also performs rate matching of the packet (e-g., by repeating or ptmcturing the data).
The interleaved packet is provided to a modulator (MOD) 216 and may be scrambled with a scrambling sequence (for an IS-95 CDMA system), covered with a chan^eli^ation code, and spread with spreading codes (e.g., short PNI and PNQ codes). The spreading witii the spreading codes is referred to as "scrambling" by the W-CDMA standard. The channelization code can be an orthogonal variable spreading factor (OVSF) code (for a W-CDMA system), a Walsh code (for an IS-95 CDMA system), or some other orthogonal code, again depeading on the particular CDMA system or standard being implemented. The £-pread data is then provided to a transmitter (TMXR) 218 and quadrature modulated, filtered, and amplified to generate one or more downlink signals. The dovmlink signal(s) axe transmitted over the-air from one or more antennas 220. The downlink processing is described in farther detail in the IS-95 and W-CDl^^A standards.

At the receiver unit ti^ dovmlink signal(s) are received by an antenna 230 and routed to a receiver (RCVR) 232. Receiver 232 filters, air^lifies, quadrature demodiilates, and samples and qxiantizes the received signal. The digitized samples are then provided to a demodulator (DEMOD) 234 and despreaded (or descrambled) with despreading codes, may be descrambled witii a descrambling code (for ihe IS-95 CDMA system), and deoovered with a channelization code for each' physical charm^ being processed- The despreading, descrambling/ and dnannelization codes correspond to the codes xised at the transmitter unit The demodulated data is lihen provided to a deoder 236 that performs the inverse of the functioxxs performed at encoder 214 (e.g., the de-interleaving, decoding, and CSC check functions). The deo>ded data is provided to a data sink 238.
A controller 240 can direct the operation of demodiilator 234 and decoder 236. A memory mtit 242 couples to demodulator 234 (and possibly also to controUo: 240, as indicated by the dashed line) and, in c&tain modes of operation, can be used to store intermediate res\iltS/ or data, from demodulator 234
The block diagram, as described above, supports transmissions of padcet data, messaging, voice, video, and other types of communication on the downlinks A bi-directional communications system also supports uplink transmission from the remote station to the base station. However, the uplink processing is not shown in FIG. 2 for simplicity.
HG. 3A is a block diagram of a modulator 300, a transmitter 302, and antennas 304 that support transmission of a physical channel over two antennas. The processing imits shown in FIG. 3A can be used to support a space time block coding transmit antenna diversity (STTD) mode defined by the W-CDMA standard. The data for the physical channel (i*e., the channd data) is provided to a STTD encoder 310 that generates STTD encoded data for each of tKe antennas used to transmit the channel data. The operation of STTD encoder 310 is described in further detail bdow. The STTD encoded data for each antenna is provided to a re^ective channelizer 320 that covers the data with a channelization code assigned to 1-Ke physical diaimel to generate . "channelized" data. In the W-CDMA system, the same channdizatiOTi code is used for both antennas.
Covering is x^sed to channelize data sudi that the data on multiple physical channds can be optimally transmitted- and received without interfering with each otiher. Each physical channel is assigned a particular channelization code sdected from a set of channelization codes. The codes in

the set are typically designed to be orfltogonal with each other such that a pariicular code multiplied with itself and integrated over the length of the code will result in a high (energy) value, but if multiplied with other codes in the set and integrated over the code length will result in a low (or ideally, zero) vahie, Nort-orthogonal channeli3ation codes can also be tised for the covering.
To perform covering, the diannel bits to be transmitted are multiplied wititi the assigned channelization^code- At the receiver, the transmitted bits are recovered by multiplying the received samples with the same code used to cov Within channdizer 320, the STTD encoded data is provided to an I/Q demultipl^er (DEMLDQ 322 that demultiplexes the data into inphase (I) and quadrature (Q) data. The I data and Q data are provided to respective multipliers 32fe and 324b and covered (i.e-, multiplied) wlti^ the channelization code, Cd, assigned to ihe physical channeL Multipliers 324a and 324b perform covering using the channelization code in STOJlar manner a$ the covering performed with Walsh code in the IS-95 CDMA system.
For an IS-95 CDMA system, Walsh codes having a fixed length of 64 chips are used to cov©: ihe traffic channels, wilh each physical channel having a varLible but limited data rate (e.g., ^ 32 Kbps), W-CDMA uses orthogonal codes called OVSF codes tihat are identical to Walsh codes except that the index identifying Ihe code is bit-reversed (e.g., for a lenglh 64 code, Walsh code (64,5) (5 = bOOOlOl) would be OVSF code (64,40) (40 -^ blOlOOO)). W-CDMA uses OVSF codes with length Ihat varies from 4 to 512 chips and is dependent on the data rate of the physical channel. OVSF codes are described in further detail below.
The covered Q data from multiplier 324b is provided to a multipliex 326 and multiplied with the complex symbol, ], to generate the imaginary part of the channelized data. The real part from multiplier 324a and the imaginary part from multiplier 326 are combined by an adder 328 to provide tiie complex channelized data. The dhannelized data for each antenna is then scrambled with a complex scrairibling code, PN, by a multiplier 328 and scaled with a weight factor, G, by a mtdtiplier 330- The weight factor, G, is sdected for the physical channel being processed and is used to adjust the transmit power of the physical channels

The scrambled and weighted data from multiplier 332b, the scrambled and weighted data for other physical diaimels, and other data for some other physical charmels (e.g./ the coinmon control physical channel) are combined by an adder 334 to generate composite data- Itie composite data for each antenna is further multiplied with a comply: weight factor, W, by a midtiplier 336. As specified in the W-CDMA standard, the weight factor is used for phase adjustment in dosed loop mode 1 and for phase/ampKtude adjustment in dos'^d loop mode 2.
The adjusted data from eadn multiplier 336 is then provided to a reqsective transmitter 302 that converts the data into an RF modulated signal, whi(ii is then transmitted from a respective antenna 304.
FIG. 3B is a block diagrara of an embodiment of a portion of transmitter 302. The composite data from modulator 300, having complex-values, is provided to a real/imaginary demultiplexer (Re/Im DEMUX) 352 that demultiplexes the data into the real and imaginary parts. The real and imaginary parts are then filtered by respective pulse shape filters 354a and 354i. The filtered real and imaginary parts are modulated with respective carrier signals cos(a>gt) and -sxn(a>jt) by .multipliers 356a and 356b to generate inphase and quadrature modulated components that are titien combined by an addCT 358 to generate a modulated signaL The modulated signal is typically conditioned prior to transmission from antenna 304.
FIG, 4A is a diagram that illustrates the encoding performed by STTD encoder 310. The diannel data comprises a sequence of bits represented as {b^ by b., by b^, b^ b^ by, ♦,.}• T^ *« STTD mode of the W-CDMA system, the input bit sequence is transmitted from two or more antennas. For example, each antenna can be substituted by a phased array antenna consisting of multiple antennas having a particular beam pattern. Hie STTD sdieme is designed for two idiverse paths (regardless of the number of physical antennas employed).
STTD encoder 310 receives the input bit sequence and generates a number of output bit sequences, one output bit sequence for each antenna xised to transmit the chaimel data (two output bit sequences are generated in the example shown in FIG- 4A). In accordance with the W-CDMA standard, the first output bit sequence for antenna 1 is a replica of the input bit sequence, and the second output bit sequence for antenna 2 contain the same bits, but the order of tihe bits is rearranged in time and some of the bits are inverted-
The STTD encoding to generate the second output bit sequence is performed by first partitioning the input bit sequence into blocks of four bits. The bit in the tiiird bit position of each block (e.g., b^ is swapped with the bit in

the first bit position (e.g., b^/ and the bit in the fourth bit position (e,gv b,) is swapped wiih the bit in the second bit position (e-gv b,). The bits in the first and foxartit bit positioits in the block (e.gv b^ and h^ are also inverted in the second output bit sequence-
FIG- 4B is a diagram that iUusfarates the encoding performed by STTD encoder 310 using complex notation. Prior to transmission, the ou^ut bit sequences from STTD encoder 310 are provided to respective trananitters- In each transmitter, the bit sequence is demultiplexed into an inphase (I) sequence and a quadrature (Q) sequence. For antenna 1, the transmitter ii^ut bit sequence \b^ b,, by by b^ b^, b^, b^,.,.} is demultiplexed into the I sequence {b^ h^ K K —) ^ Similarly, for antenna 2, the transmitter input bit sequence f b^, by b^ -bj, -b^ IV K "^5^ •'*} ^ demultiplexed into &e I sequence {-ty b^^ -b^ b^,...} and the Q sequence {by -bj, b,, -by .- ). The I and Q sequerures for antenna 2 can be vie^f^d as a complex symbol sequence {-Si*, S^*, -Sj*, S^*, ►..}, where -S^* ^ -b^ -^ jby £/ -he' py and so oru S/ represents the complex conjugate of S^.
Transmission of the channel data over two or more antennas provides spatial diversity, which can improve performance against deleterious path effects- Rearranging the order of the bits in at least one of the transmissions provides temporal diversity, which can improve performance against impulse nois(i axwi interference- The bit inversion is used to allow the receiver unit to separate out the streams from the two antennas and consequently take advantage of the path diversity.
In accordance with the W-CDMA standard, the downlink shared channel (DSCH) can be used to transmit data at a variable bit rate ranging from 15 Ko^s up to 1-92 Mbps. For the W-CDMA standard; the bit rate refers to the rate of the bits provided to modulator 216, i-e., after encoding, interleaving, and rate matching but brfore covering, spreading, and so oru The OVSP codes oised to cover (and channdize) the data has a fixed chip rate (Le., 3.84 Mcps for W-CD^^IA) but a variable length ranging from four chips to 512 chips* The length of the OVSF code is also referred to as the spreading factor, SF» Table 1 lists the bit r£.tes (in Kbfps) supported by W-CDMA and the correspcmding OVSF code lengthis (in chips).



As iihovm in Table 1, the OVSF code has a maximxtm length of 512 chips for the lowest supported bit rate of 15 Kbps and a ixiinimxim length of 4 chips for the hig^\est supported bit rate of 1.92 Mbps.
FIG, 5 is a diagram that illustrates the generation of the OVSF codes. Each OVSP code can be identified by the designation/ 0^,^, where SF is the spreading factor (which is equal to the length of the code) and i is the identity of tlie particular code (Le., i = 0,1, 2, •.. SF-1) for that spreading factor/ The OVSF codes are "structured" codes, and successively longer OVSF codes can be gen*a:ated from shorter OVSF codes in accordance with a set of rules. To gen As shown in FIG- 5, only one OVSF code is defined for a code length of one (i.e., C^ == 1). Two OVSF codes are defined for a code length of two (Lev C^ = 1;L and C^ = 1,-1)/ and are generated from the OVSF code of length 1 (i.e., Ci^)- Similarly, four OVSF codes are defined for a code length of four, with the OV5F codes (C,^ = 1,1.U) and (C^^ = IX-lrl) being generated from the OVSF code (C^ = 1,1) and the OVSF codes (C^ = 1,-1,1,-1) and (Q^ = 1,-1,-1,1) being genfcjrated from the OVSF code (Cy = 1,-1).
In accordance with the W-CDMA standard, each user in communication with the base station is typically assigned a dedicated channel (DCH) on the downlink and may be aissigned/allocated a downlink shared channd ,(E)SCH) as needed for packet data transmisdon. The parameters for the dedicated chanjoel are either negotiated or may be known to the receiver unit at the time com:nunication is established* For example, for the dedicated channel, the

dhaimeiization code is known but other parameters such as rate inatehing paiameters may or may not be known- If the parameters are not known at the time commtmication is established, a transport format combination indication (TFCI) is used to indicate the unknown tran^ort format
However, in part because of the dynamic nature of the downlink shared channel, its channelization code is typically not known a priori but provided instead on the dedicated chaifeiel. The transport format indudes various parameters associated with the downlink shared channel such as, for example, the particular OVSF code asagned to the channel, the bit rate, and so oa The transport format of the downlink shared channel and the transport formats of other physical channels axe combined into a transport format combination (TFC) and encoded into a TPCI that is transmitted to the receive unit on the dedicated channel. The TPQ and its coding and transmission are described in debiil in the W-CDMA standard.
Upon detection of the TPQ in the dedicated channel, the receiver unit is able to determine the transport formats for the individual physical channels, including that of the downlink shared dxannd. The receiver unit is then able to deniodulate and decode the downlink shared channel.
In some instances, ^e TFQ is transmitted on the dedicated channel approximately concurrently with the data transmission on the downlink shared chaainel. Furthermore, because of itie processing delay to demodulate and dea>de the dedicated channel, the TFQ may not be known until a short time after reception of ihe data transmission on the shared channel. The fccaixsport fonaiat information necessary to demodulate and decode ti:is: shared channel maj^ therefore not be available in real-time. The invention provides techniques to efficiently process the data transmission when the actual OVSF code is not known.
FIG. 6 is a flow diagram of ttie processing of a modulated (spread spectrum) signal in accordance with an embodiment of tiie inv^:ition. Initially/ the modulated signal is received,, conditioned, and digitized to generate samples, at step 612. The samples may thereafter be processed by a nxuhbex of fingsr elements, with eadi finger element assigned to process a particular signal . path. (i.e./ multipath) of the received signal.
For each finger element, the samples are despread with a FN sequence at a time offset corresponding to the multipath being processed, at step 614. A hypothesized OVSF code is then identified for use in decovering the despread samples, at ^tep 616, The hypothesized OVSF code is a code that can be iised to generrate the actual OVSF code used to cover the data at the base station, and is

descTibed in further detail below. If iiie received signal is from more tihai> one bas€! station and different actual OVSF codes are used at the transmitting base staticOTS/ then different hypothesized OVSP codes may be used for different fingsr dements.
For each finger element, the despread samples are then decover with the hypodtesized OVSF code, and may further be coherently demodulated with pilot estimates, if available, to generate pardaiiy processed symbols, at step 618. The partially processed symbols represent a portion of the actual symbol or the entire actual symbol, depending on the relationship between the hypothesized and actual OVSF codes, as described in further detail below. The partially processed sjntribols from all assigned finger elements are then appropriately weig;hted and combined to generate combined synibols, or intermediate results, at st€p 622. The intermediate results may be stored to a buffer until the actual OVSF code is known.
At step 624, the actual OVSF code is determined- The intermediate resvilts may then be retrieved from the buffer and further processed. Specificsilly, the intermediate results for each symbol period (Le., each actual OVSF code length) are scaled and corhbined to generate the final result for that symbol period, at step 626. The scaling is dependent on the relationship between the hypothesized and actual OVSF codes, as described bdow. The final result represents an estimate of the transmitted sjmnibol. For STTD decoding in the W-CDMA system, the final results from multiple symbol periods are scaled and combined, in a complementary marmer to the STTD encoding performed at the base station^ to generate recovered symbols, at step 628. The processing shown in FIG* 6 is described in further detail bdow.
FIG. 7 is a block diagram of an etribodiment of a portion of a receiver unit TOO that can be used to receive and demodulate a physical doannel, including one transmitted from multiple transmit antennas in 1he STTD mode of the W-CDMA standard. One or more RF modulated signals from one or more transmit antennas are received by an antenna 710 and provided to a receiver ^CVR) 712 that cx)nditions (e.g., air^iifies, filters, and so on) the received signal and quadrature dovraconverts tihe conditioned signal to an * intermediate frequency (IF) or baseband. Receiver 712 also samples and quantizes the downconverted inphase and quadrature signals to generate received samples that are then provided to a rake receiver 720. Although a rake receiver is shown in HG. 7 for processicig the physical channel, other receiver striictures and implementations can also be used and are wilhin the scope of the preseat invention.

In typical implemeatations, the received signal is sampled at a sample rate, f^, that is higher liian the chip rate, f^, of the received signaL For exart^le, the diip rate may be fc »12288 Mcps for the IS-95 CDMA system (or 3 for the W-CDMA system) but the sample rate may be, for example, 8 times (i.e., Bxddp), 16 times (i-e., 16xchip), 32 times (i.e., 32xd:up), or other multiple of the chip rate. The higher sainple rate allows for fine adjustment of the timing to "zoom in" on a path position.
As shown in HG. 7, rake receiver 720 includes a searcher element 722 and a number of finger elemente 730a through 730n. Each of &ese elements receives the samples from receiver 712 and performs the tasks associated with the (dement or as directed by a controller 740. For example,, searcher element 722 may be instructed by controller 740, or assigned to search for strong
. instsinces of the received signal. The strong signals may be present at different time offsets, and can be ideatijaed by searcher element 722 by processing Ihe sam.ples with different parameters (e-g., different FN codes, different time offsets, and so on). Searcher 722 may be deagned to provide data corresponding to ti^e searched signal or an indication of the search result to contx)Iler 740. Controller 740 assigns §nger elements 730 to demodulate Ihe
' strongest instances of the received signal, as determined with the assistance of searciher element 722.
Each assigned finger element 750 performs demodiolation of one physical channel for one instance of the received signal (i,e*, a signal at a particular assigned time offset), as directed by controller 740. When the channelissation code is knowiv each asagned finger element 730 provides recoA'-ered sytribols (e-g., SJ coixesponding to the assigned instance of the received signal The recovered symbols from aH assigned finger dements 730 are then provided to a combiner 732 and combined to provide composite symbols that are more indicative of the transmitted data. The combined symi^ols represent the recovered channel data, and are provided to the subsequent processing block (e^g., the decoder).
In accordance with the invention, in instances where flie channelization code is not known, each assigned finger dement 730 provides partially processed symbols corresponding to the assigned instance of the received signal (i-ev the assigned signal paih). Each partially processed syinbol can represent one recovered symbol or a fraction of one recovered symbol,
depeaading on a hypothesized channelization code used to process tiie received
«
samples and the actual channelization code- The partially processed symbols from various assigned finger dements 730 may also be combined to provide

combined symbols more indicative of iiie transmitted data. The combined symbols from the combiner represent "intermediate results" tiiat can be pro-^rided to, and stored by, a memory 742. The intermediate res\ilts axe later retrieved and may further be processed, when the actual channelization code is kno'wn, to provide ttie recovered symbols representative of the recovered channel data. Processing with the hypothesized chaimeHzation code and postprocessing of the intermediate data eure described in further detail below.
The design and operation of a rake receiver for an CDMA system is described in further detail in VS. Patent No. 5,764,687, entitled "MOBILE DEMODULATOR ARCHTTECTURE FOR A SPREAD SPECTRUM MULTIPLE ACCrESS COMMUNICATION SYSTEM," and U-S. Patent No. 5490,165, entitled "DEMODULATION ELEMENT ASSIGNMENT IN A SYSTEM CAI*ABLE OF RECEIVING MULTIPLE SIGNALS," both assigned to the assij;nee of ihe present invention and incorporated herein by reference.
Although not shown in FIG. 7 for simplicity, each finger element can also indude a lock detector that computes a quality indicator (e-g., an average energy) of the recovered data at the finger element and masks the partially processed symbols from the finger dement if the quality indicator does not ■ exce-sd a minimum threshold. The masking ensures that only finger dements receiving signal of sufficient strength and reliability will contribute to the coml:)ined output, thxis enhancing the quality of the recovered data-
FIG. 8 is a block diagram of an embodiment of a finger dement 800 that can be used to implement one finger element 730 in FIG. 7. The fingsr dement is abo referred to as a demodulation dement. Ihe received samples from receiver 712 are provided to a multiplier 812 and descrambled with a complex desorambling code, PN, corre^onding to the scrambling code used at the transmitter unit and having a time offset assigned to the finger dement The desarambled samples are then provided to a set of multipliers 814a through 814c
The descrambled saaiqples indude data for all physical channds in the recdved signal, induding pilot data, control data, and data transmitted on the dedicated and shared diannds. Separation of these various types of data is adueved at lihe transmitter unit by (1) dianneliging eadi digumd da^. with an OVS? code assigned for the particular diannd, (2) time multiplexing the data, or (3) a combination of both.
In accordance with the W-CDMA standard, for dosed loop mode 1, orthogoxtal pilots are used for the two transmit antennas. The orthogonal pilots for tirie two antennas are generated by using different symbol pattemj, W^^ and

Wpj. Thus/ at the receiver uiut, tiae pilots are recovered by irtultipliers 814a and
814t) using the same symbol patterns, W^^ and W^,. The pilot samples from
multipliers SlAa and 814b are then provided to respective accumulators S16a
and $l6b and accumtdated over the lengiii of the syihbol patterns to obtain
values indicative of the instantaneous phase and amplitude of the pilots. The
instantaneous pilot values from accumulators 816a and 816b are thea provided
to respective pilot processors 8lSa and 818b.
The recaved pilot data is filtered and processed (e.g., iivtepolated ox
extrapolated, depending on the type of ^gnal processing enr^^loyed) to gentjrate pilot estimates, ^ and ^, used to demodulate ihe data. The de^read
pilot samples may also be provided to a pilot detector PET) 822 that provides an bidication of the quality of ihe received pSots. In a spediic implementation, pilot detector 822 measures the power of the received pilots and provides a pilot quaHty estimate, ?.
In similar manner, the descranibled samples from multiplier 812 are provided to a multiplier 814c and decovered wilh. either the actual OVSF code, C^ (if it is known) or a hypotheazed OVSF code, C^, (if the actual OVSF code is not ioiown) to recover the data on the physical channel. The samples from multiplier 814c are then provided to an accumulator 816c and accumulated over the length of the OVSF code, C^ or Q, used to decover tine samples. If the actujil OVSF code, C^ for the physical channel being processed is known, acojtnulator 816c accumvdates the samples over the length of the code to generate a received sytnboL For exdmple, referring back to Table 1, if the bit rate of the physical chaimel is 1^92 Mbps, accumulator 816c accumulates the samples over a four-chip period to provide a received ^ronbol. On the other extreme, if the bit rate of the physical chaimel is 73 kbps^ accumulator 816c accumulates the samples over a 512-chip period to provide a received symbol-The decovered symbols from accumulator 816c may be delayed by a delay elemtmt 8^ to match the delay of pilot processors 818.
In accordance with the inventiarv when the actual OVSF code is not
knov;Ti, the decovering is performed using ihe hypothesized OVSF code, Q,
having a spreading factor that may be less than that of ^e actual OVSF code,
The accumulation by accumulator 816c is then performed for a time period
correi^onding to the length of the hypothesized OVSF code. Thus, each
decox-ered symbols from accumulator 816c may be only a portion of an actual
received symbol.
The pilot estimates, ^ and ^, from pilot processors SlSa and 818b and
the discovered symbols from delay dement 828 are provided to a data recovery

element 830 tiiat performs the necessary computations to generate the partially
processed symbols for the finger element. In particular/ data recovery element
830 may perform a dot prodtict and a cross product between the pilot estimates
and the decovered symibols to generate the partially processed symbols. The dot and cross products refer to the operations XjPj + X^P^ and X^P^ - X^Pj,
resj>ectively, where the I and Q components of the decovered data symbol are resj>ecfcively X, and X^ and the I and Q components of the estimated pilot are resp>ectively Pj and P^ Hie data recovery process may further involve rotation and/or scaling of either or both of the decovered data symbol and estimated pilct, as is known to those skilled in the art.
Ixi FIG. 8, multipliers 814a and 814b, accumulators 816a and 816b, and pilot processors 818a and 818b may be referred to as the pilot processing unit of the finger element For a mode in which pilot data is transnutted from only one anteaina, only one of the pilot processing paths is needed and the other pilot processing path can be disabled- Multiplier 814c accumulator 816c> and delay element 828 may be referred to as the data processing unit of the finger element.
The physical channel being processed may be transanitted from one antenna or two antennas (e-g., for the STTD mode in the W-CDMLA. system). Moreover, the physical channel may be modxalated using either BPSK modulation (e-gv for an IS-95 CDMA system) or QPSK modtilation (e-g., for a W- In the STTD mode of the W-CDMA system, two RF modulated signals are transmitted from two trananit antennas for one physical channel. Each of the ".RF modxjlated signals can experience independent and different path loss. The received signal is thus a weighted sum of the two RF modulated signals, and can be expressed as:

where X is the received symbol sequence {X^^, Xj, Xj, Xy .-.}, S^ is the symbol sequience (S^ S,, S^, S^, —) transmitted from the first transmit antenna, S^ is the , symbol sequence {-Sj*, S/, -Sj*, Sj*, ...} transmitted from the second transmit antenna, and a and P are the fading coefficients for the two paths. Hie received symibols can be expressed as:



EquciUons 1 ihrough 4 assume that ihe signals transirutted from antennas 1 and 2 are received at the finger element at approximately equal time, which is generally true in many instances- If the two paths are not received at the same time, then there may be some "small degradation in performance, but the equations give above are still valid and the processing described herein can still be applied.
Because of time mTJ.tiplexing of the sjnnbols transmitted on the second traniimit antenna, the received symbol sequence Pv Xj, Xy Xj, -..} are further processed to recover the transmitted symbol sequence {S^ Sy S^ Sy ...}. To dete::mine an estimate of the transmitted symbol S^, equations 1 and 2 may be scaled and con^ibined as follows:

where JQ is tihe recovered symbol/ IS and ^ are complex-valued estimates of
the }>ath losses (i-e., channel or fading coefficient) from the first and second
transmit antennas^ respectively, to the receive antenna- The noise term is due
to iiriperfect channel estirctation. Since the pilots are also transmitted from both
trans-mit antennas, and the pilot data value is known, the pilot estimates desaibed above can be used as the channel estimates, a and ^, in the above
equations. Similarly, to determine an estimate of the transmitted sjmbol S^, equations 1 and 2 may be scaled and combined as follows:



To recover the daia transmitted an fhe physical channel, it is necessary to decover the descrambled samples with the same OVSF code, C^, iised to channelize ihe physical diannel at the transinitter* However, as noted above, the bit rate for the shared channel in the W-CDMA system is variable and the bit rate for a particular radio frame is provided on the dedicated channel, which may be transmitted approximately concurrOT.tiy with the data*
Several techniques can be used to recover the data on a physical channel (such as the downlink shared channel in tbie W-CDMA system) having a variable but imlcnown bit rate (Lev imknown OVSF code). Ixi one technique, the d-esaambled samples are stored to memory and later retrieved for processing when information on ihe physical channd is kxu>wru Becaiise of varioxis factors (e.g., the size of the radio frame, the hi^ bit rate, the higher resolution used to represent the unprocessed samples, and so on), a large memciry is required to store the unprocessed samples* Also, descrambled samples are stored for all assigned finger elements.
In another technique, the descrambled data can be decovered for all possitde OVSF codes (or all OVSF codes possible with ihe physical channel being processed). For example, the descrambled data can be decovered for OVSF code C^ = lAAA/ OVSF code C^^ = 1,1,-1,-1/ and so on, conduding vri&i OVSF code C^^^^ = 1,-1,-1,1,-1,1,1,-1,-. Since there are almost 1024 pos^le OVSF codes, the brute force processing for all possible OVSF codes can be extensive and the implementation cost may be high Thus, ottier techniques that can reduce the amount of necessary buffering and/or computation are highly desirable.
In accordance with the invention, if Ihe actual OVSF code is not known, a hypothesized OVSF code is used to process the received santples. The actual OVSF code is the code that was actually used to channelize the physical channd being processed. Initially, a set of ail OVSF codes that can be used to generate all possible actually used OVSF codes is determined. The OVSF code wilh the largest ^reading factor within this set can then be selected as the hypotJiesized OVSF code. The hypothesized OVSF code is thus a 'T?ase" OVSF code tiiat can be used to generate all possibilities of the actual OVSF code. The code tree for generating the OVSF codes has a known structure and this propeity of the OVSF code is exploited in the present invention. If the actual

OVSF code is known to be a meiiiber of a set of possible codes, then the hypothesized OVSF code with the largest spreading fector that can be used to generate the possible codes in the set is tised to decover tlie data.
For example, referring back to FIG- 5, if the actual OVSF code has a sprcsading fector of 4 and it is known that the code is either C^ or C^, then the OVSF code C^ can be used to process the physical channel since (Z^ can be used to generate C^^ and C^. As another example, if the actual OVSF code is known to be a member of a set that includes C^^ to C^^j and Q^^ to C^^, then the OVSF code C^ can be used to process the physical channel. If the actual OVSF code is known to be a member of a smaller code set that includes C^^ to C,^ and C^ to C^^, then the OVSF code C^ having a larger ^reading factor can be used to process the physical channel. The OVSF code selected for use to process the phyiyical channd is referred to as the hypothesized OVSF code.
For a dearer understanding of the invention, the processing of the received samples using a hypothesized OVSF code is now described for a specific example. In this example, the hypothesized OVSF code Is half the length of the actual OVSF code.
FIG. 9 is a diagram that iUxistrates the processing of two sym&ols using the actual OVSF code and the hypothesized OVSF code. In this example, the filtex'ed pilot received from the first antenna is denoted as o^ tiie filtered pilot received from the second antenna is denoted as /?, the received symbols processed using the actual OVSF code are denoted as X^ and X,, and the received s3nchbols processed using the hypotheazed OVSP code are denoted as Y^ y„; y,y and y„. Since the hypothesized OVSF code is half the length of the actual OVSF code, two symbols Y^ and Xd are generated for eadi symbol X^.
If the actual OVSF code is known, ihe following terms can be computed for eiich finger element assigned to process a particular signal path:



wheire the syiribol" " denotes the weighted sum for a particular term (e.g., a^X^) across all assigned finger elements. For the STTD mode in the W-CDMA
system/ the traaismitted symbols S^ and Sj can then be determined as:

However, if the actrtal OVSF code is not known, a hypothesized OVSF code can be generated and used for processing ihe received samples. For each assijpaed finger element, the following terms can be computed xising the hypothesized OVSF code:

Since the actual OVSF code is twice the length of the hypothesized OVSF code in tihis example, two pairs of partially f^ocessed symbols are generated for each transmitted symbol. The terms from all assigned finger elements can be appi'opriately weighted and combined to obtain the following terms:

The above terms represent intermediate results that can be stored in a merriory titie actual OVSF code is detenrdned- Once the actual OVSF code is knovm, the intermediate results can be retrieved from memory and appropriated combined as foUows:


In tiie above example^ a plus sign ("+") is ^^sed to combine the two "Y" terms to generate each 'X* term. Howevar^ the "Y" terms inay be scaled differently prior to the combining^ with each 'T' term being scale witi\ either a +1 or -1 depending on how the hypothesized OVSF code is scaled and coiicatenated to obtadn the actual OVSF code. The transmitted symbols Sp and S, can then be det€'rmined in similar manner described above, as follows:

As illustrated by the above example, equations (5) and (6) cannot be executed in its entirety with the hypothesized OVSF code becaoise the lesagth of the i^yiribols S^ and S, is not known. For example, if the hypothesized OVSF coder has a length of 4 chips and the actual OVSF code has a length of 8 chips, then each transmitted symbol lasts two hypothesized OVSF codes. As another example, if the actual OVSF code has a length of 16 chips, then each transmitted s5rml>ol lasts four hypothesized OVSF codes. With the invention, partial processing is performed using the hypothesized OVSF code and the intermediate results are later combined to provide the desired outputs.
For the simple example described above, the same pilot estimates, a and >5, are used for all four symbols Y^ YQ,, Y^^ and Y^, However, it is also possible to uije the most updated pilot estimates available at the time of the symbol being processed. In this case, instead of using a single set of ^and A four sets a^ Bxxd p^ through a;, and p^ can be \ised to process the symbols Y^ through Yjj, r*5spectively.
Referring back to FIG. 8/ each finger element processes one instance of the received signal to generate the partially processed syiribols for that finger. Specifically/ each finger de^reads the received samples, decovers Ihe despread samples using the hypothesized OVSF code, and demodulates the decovered symbols with the pilot estimates to generate the partially processed symbols. The sictual OVSF code may be made up of a concatenation of Q appropriately

weigiated (with eitixer -fl or -1) hypoiiiesized OVSF codes, depending on the pattern of the actual OVSF code. Thus, Q pairs of partiaJly processed symbols are generated by each finger element for each transmitted symbol. For the W-CDMA system, Q can be 1,2,4,..., or 128, depending on relationship between the actual and hypothesized OVSF codes. Each pair of partially processed symbols thus corresponds to a portion of the transmitted symbol (for Q = 2,4, .,. or 128) or tfie entire transmitted symbol (for Q - 1). Each finger element computes the Q pairs of partially processed sjTnbols as (cfTj and ip^*), with
one pair of partially processed symbols generated for each hypottiesized OVSF code togth.
For data transmission received from one base station, the same hypoiiiesized OVSF code is used for all assigned finger elements. As noted above, for data transmissions received from mxaltiple base stations and covered with different actual OVSF codes, different hypothesized OVSF codes may be used to decover the received samples, For both cases, the partially processed symbols from all assigned finger elements are combined to generate the intermediate results.
The partially processed symbols from all assigned finger elements are then provided to combiner 732 and appropriated weighted and combined into comb:ined symbols, which represent the intermediate results. Spedfically, the partially processed s3riribols from all assigned finger elements for each hypothesized OVSF code length are scaled and combined to generate the interniediate results for that hypothesized OVSF code length. Again, Q pairs of intermediate results, (OTTQ ) and \^j through p*^) and [jS^Q^^jr are
gener:ated for each transmitted symbol. The intermediate resialts are then provi(ied to the next processing unit or stored.
Once the actual OVSF code is known, the intermediate resiilts can be retrieved and further processed, if necessary, to obtain the recovered symbols. As an example, if the physical channel was processed using a hjrpothesized OVSF code C^ = 1,1 and the actual OVSF code is C^ = 1,1,-1,-1, then the interaiediate results are retrieved, appropriately inverted as defined by the actual OVSF code, and integrated over the length of ttie actual OVSF code to obtain the recovered symbols. For the above example, the second pair of intermediate results is inverted to account for the inversion in the second half of the C^^ code. The first set of intermediate results, corresponding to (prT^j
through |ar*yg-i )r 6^c>m the Q pairs are combined to obtain the term \(X*Xj. Similarly, the second set of intermediate results, corresponding to (^V)

through \^^-i*)/ from the Q pairs are combined to obtain the term ^*).
lliese terms axe then combined as shown in equations (5) and (6) to obtain the recovered sj^mbols. Referring to FIG. 7, the processing of the intermediate resitlts can be performed by controller 740, by circuitry 'within combiner 732, or by other circuitry not shown in FIG. 7*
The processing imits, described herein (e.g., multipliers 814, accumulators 816, pilot processors 818, data recovery element 830, controller 740/ and others) can be implemented in various manners such as an application specific integrated circuit (ASIC), a digital signal processor, a miaocontroUer, a miCTOprocessor, or other electronic drcuits designed to perform the functions desciibed herein. Also, the processing tmits can be implemented with a gentaral-pttrpose or spedally designed processor operated to execute instixiction codes that achieve the functions described herein. Thus, the processing units descaibed herein can be implemented using hardware, soft^vare, or a combination thereof.
Itie memory unit can be implemented memory technologies induciing, for €'xample, random access memory (RAKf), Flash memory, and others. The memory unit can also be implemented with storage element such as, for example, a hard disk, a OD-ROM drive, and others. Variotzs other implementation of the memory unit are possible and within the scope of the presimt invention
The foregoing description of tihe preferred embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to othei- embodiLnents without the use of the inventive faculty. Thus, the present invention is not Intended to be limited to the embodiments shown herein but is to hii accorded the widest scope consistent with tihe principles and novel features disclosed herein-




WE CLAIM :
1. A method for recovering data transmitted on a physical channel, wherein a channelization
code used for the physical channel is not known at the time of the data recovery, the method
comprising the steps of:
receiving and processing a modulated signal to provide received samples;
selecting a hypothesized channelization code for processing the physical channel;
processing the received samples with the hypothesized channelization code to generate partially processed symbols;
storing intermediate results representative of the partially processed symbols;
determining an actual channelization code used for the physical channel;
processing the intermediate results in accordance with the actual channelization code and the hypothesized channelization code to provide fmal results, wherein the processing the intermediate results includes
partitioning the intermediate results into sets of intermediate results,
scaling each intermediate result in a particular set with a respective scaling factor determined by the actual channelization code and the hypothesized channelization code, and
combining the scaled results for each set to obtain a final result for the set.
2. The metiiod as claimed in claim 1, wherein the processing the received samples includes
decovering the received samples with the hypothesized channelization code to generate decovered symbols, and
demodulating the decovered symbols with pilot estimates to generate the partially processed symbols.
3. The method as claimed in claim 2, wherein the demodulating with the pilot estimates
includes
performing a dot product between the decovered symbols and the pilot estimates, and performing a cross product between the discovered symbols and the pilot estimates, and wherein the partially processed symbols are derived based on results of the dot and cross products.

4. The method as claimed in claim 2, wherein the pilot estimates are generated by despreading the received samples with a pilot despreading code, and integrating the despread pilot samples over a length of the pilot despreading code to obtain pilot symbols that are then used to generate the pilot estimates.
5.The method as claimed in claim 4, wherein the pilot estimates are generated by interpolating or extrapolating the pilot symbols.
6.The method as claimed in claim 1, further comprising the steps of:
combining partially processed symbols from a plurality of demodulation elements assigned to process the physical channel to generate the intermediate results.
7. The method as claimed in claim 1, wherein the scaling factor is either +1 or -1.
8. The method as claimed in claim 1, wherein the hypothesized channelization code is a member of a set of channelization codes that may be used to generate the actual channelization code, and wherein the hypothesized channelization code has a length that is shorter or equal to that of the actual channelization code.
9. The method as claimed in claim 8, wherein the hypothesized channelization code can be used to generate all channelization codes in the set.

10. The method as claimed in claim 8, wherein the hypothesized channelization code is an orthogonal variable spreading factor (OVSF) code.
11. The method as claimed in claim 10, wherein the hypothesized OVSF has a largest spreading factor among the channelization codes in the set.
12. The method as claimed in claim 10, wherein the hypothesized OVSF code has a spreading factor of four or greater.

13. The method as claimed in claim 10, wherem the hypotnesizea uv^r coue miu tui tu^iuai OVSF code each has a spreading factor ranging from four to 512.
14. The metiiod as claimed in claim 1, wherein the physical channel has a variable data rate.
15. The method as claimed in claim 1, comprising the steps of:
selectively combining the fmal results from multiple symbol periods to obtain a recovered symbol, wherein each symbol period corresponds to duration of the actual channelization code.
16. The method as claimed in claim 15, wherein the selectively combining is performed in a manner complementary to an encoding performed in accordance with a space time block coding transmit antenna diversity (STTD) mode.
17. A receiver unit operative to process a physical channel in a CDMA communications system, comprising the steps of:
a receiver 232 operative to receive a modulated signal and provide received samples indicative of data transmitted on the physical channel;
at least one demodulator element 234 coupled to the receiver, each demodulator element including a data processing unit operative to receive and process the received samples in accordance with a hypothesized channelization code to provide decovered symbols;
a memory unit 242 operative to store intermediate results representative of the decovered symbols from the at least on demodulator element; and
a processor operative to receive and process the intermediate results in accordance with an actual channelization code and the hypothesized channelization code to generate final results, wherein each demodulator element further includes:
a pilot processing unit operative to receive and process the received samples to generate pilot estimates, and

a data recovery element 830 coupled to the pilot processing unit and the data processing unit, the data recoveiy element operative to receive the pilot estimates and the decovered symbols and generate partially processed symbols; and wherein the processor processes the intermediate results by partitioning the intermediate results into sets of intermediate results, scaling each intermediate result in a particular set with a respective scaling factor determined by the actual channelization code and the hypothesized channelization code, and combining the scaled results for each set to obtain a fmal result for the set.
18. The receiver unit as claimed in claim 17, wherein the data recovery element is operative to receive and demodulate the decovered symbols with the pilot estimates to generate the partially processed symbols.
19. The receiver unit as claimed in claim 18, comprising:
a combiner 732 coupled to the at least one demodulator element and operative to receive and combine partially processed symbols from one or more assigned demodulator elements to generate the intermediate results.


Documents:

337-chenp-2003-abstract.pdf

337-chenp-2003-assignement.pdf

337-chenp-2003-claims duplicate.pdf

337-chenp-2003-claims original.pdf

337-chenp-2003-correspondnece-others.pdf

337-chenp-2003-correspondnece-po.pdf

337-chenp-2003-description(complete) duplicate.pdf

337-chenp-2003-description(complete) original.pdf

337-chenp-2003-drawings.pdf

337-chenp-2003-form 1.pdf

337-chenp-2003-form 26.pdf

337-chenp-2003-form 3.pdf

337-chenp-2003-form 5.pdf

337-chenp-2003-other documents.pdf

337-chenp-2003-pct.pdf


Patent Number 209621
Indian Patent Application Number 337/CHENP/2003
PG Journal Number 50/2007
Publication Date 14-Dec-2007
Grant Date 05-Sep-2007
Date of Filing 04-Mar-2003
Name of Patentee M/S. QUALCOMM INCORPORATED
Applicant Address 5775 Morehouse Drive, San Diego, California 92121
Inventors:
# Inventor's Name Inventor's Address
1 TERASAWA Daisuke 10754 Chinon Circle San Diego, California 92126
2 AGRAWAL Avneesh 7891 Doug Hill, No. 29, San Diego, California 95125
PCT International Classification Number H04B 7/26
PCT International Application Number PCT/US2001/027626
PCT International Filing date 2001-09-05
PCT Conventions:
# PCT Application Number Date of Convention Priority Country
1 09/655,609 2000-09-06 U.S.A.