Title of Invention	"A CHANNEL TRANSMISSION DEVICE AND METHOD FOR CDMA COMMUNICATION"
Abstract	A channel transmission device for a code division multiple access (CDMA) communication system using at least two carriers. In the channel transmission device, a channel encoder encodes channel data to be transmitted into symbols at a predetermined coding rate. A channel controller generates a symbol distribution signal according to a predetermined symbol deleting matrix pattern. The symbol deleting matrix pattern is so determined as to distribute the symbols to the respective carriers with a minimized performance degradation even though a specific carrier is damaged. A symbol distributer receives the symbols and distributes received symbols to the carriers according to the symbol distribution signal.

Title of Invention

"A CHANNEL TRANSMISSION DEVICE AND METHOD FOR CDMA COMMUNICATION"

Abstract

A channel transmission device for a code division multiple access (CDMA) communication system using at least two carriers. In the channel transmission device, a channel encoder encodes channel data to be transmitted into symbols at a predetermined coding rate. A channel controller generates a symbol distribution signal according to a predetermined symbol deleting matrix pattern. The symbol deleting matrix pattern is so determined as to distribute the symbols to the respective carriers with a minimized performance degradation even though a specific carrier is damaged. A symbol distributer receives the symbols and distributes received symbols to the carriers according to the symbol distribution signal.

Full Text	PRIORITY This application claims priority to an application entitled "Device and Method for Generating and Distributing Coded Symbols in CDMA Communication System" filed in the Korean Industrial Property Office on May 30, 1998 and assigned Serial No. 98-2014U, BACKGROUND OF THE INVENTION 1, Field of the Invention The present invention relates generally to a data transmission device and method for a CDMA communication system, and in particular, to a device and method for generating and distributing symbols capable of preventing degradation of a channel performance during data transmission. 2. Description of the Related Art At present, code division multiple access (CDMA) communication systems are implemented based on the IS-95 standard. However, with the development of communication technology, subscribers to the communication services increase greatly in number. Therefore, there are proposed many methods for meeting the subscriber's increasing demands for the high quality service. An approach to such methods includes a method for improving a forward link structure. For an improved forward link structure, there is a forward link fundamental channel designed for a third generation multicarrier CDMA system proposed in the T1A/E1A TR45,5 conference A forward link structure for a multicarrier CDMA communication system is illustrated in FIG. 1. Referring to FIG. 1, a channel encoder 10 encodes input data, and a rate matcher 20 repeats and punctures symbols output from the channel encoder 10. Here, the data input to the channel encoder 10 has a variable bit rate The rate matcher 20 repeats and punctures the coded data bits (i.e., symbols) output from the channel encoder 10 so as to match symbol rates for the data having the variable bit rate. A channel interleaver 30 interleaves an output of the rate matcher 20. A block interleaver is typically used for the interleaver 30, A long code generator 91 generates a long code which is identical to that used by the subscriber. The long code is a unique identification code for the subscriber. Thus, different long codes are assigned to the respective subscribers. A decimator 92 decimates the long code to match a rate of the long code to a rate of the symbols output from the interleaver 30. An adder 93 adds an output of the channel interleaver 30 and an output of the decimator 92 An exclusive OR gate is typically used for the adder 93 A demultiplexer 40 sequentially multiplexes data output from the adder 93 to multiple carriers A, B and C. First to third binary-to-four level converters 51-53 convert signal levels of binary data output from the demultiplexer 40 by converting - 2 - input data of "0" to "+l" and input data of "V to "-\". First to third orthogonal modulators 61-63 encode data output from the first to third level converts 51-53 with corresponding Walsh codes, respectively. Here, the Walsh codes have a length of 256 bits. First to third spreaders 71-73 spread outputs of the first to third orthogonal modulators 61-63, respectively. Here, QPSK. (Quadrature Phase Shift Keying) spreaders can be used for the spreaders 71-73. First to third attenuators (or gain controller) 81-83 control gains of the spread signals output from the first to third spreaders 71-73 according to corresponding attenuation signals GA-GC, respectively. Here, the signals output from the first to third attenuators 81-83 become different earners A, B and C. In the forward link structure of FIG. 1, the channel encoder 10 having a coding rate of R=l /3 encodes the input data into 3 coded data bits (i.e., code words or symbols) per bit. Such coded data bits are demultiplexed to the three carriers A, B and C after rate matching and channel interleaving. The multicarrier CDMA communication system of FIG. 1 can be modified to a single carrier CDMA communication system by removing the demultiplexer 40 and using only a set of the level converter, the orthogonal modulator, the spreader and the attenuator. FIG. 2 is a detailed diagram illustrating the channel encoder 10, the rate matcher 20 and the channel interleaver 30. In FIG. 2, data of a first rate is composed of 172 bits (full rate) per 20ms frame; data of a second rate is composed of 80 bits (1/2 rate) per 20ms frame; data of a third rate is composed of 40 bits (1/4 rate) per 20ms frame; and data of a fourth rate is composed of 16 bits (1/8 rate) per - 3 - Referring to FIG. 2, first to fourth CRC generators 111-114 generate CRC bits corresponding to the respective input data having different rates and add the generated CRC bits to the input data. Specifically, 12-bit CRC is added to the 172-bit data of the first rate; 8-bit CRC is added to the 80-bit data of the second rate; 6-bit CRC is added to the 40-bit data of the third rate; and 6-bit CRC is added to the 16-bit data of the fourth rate. First to fourth tail bit generators 121-124 add 8 tail bits to the CRC-added data, respectively. Therefore, the first tail bit generator 121 outputs 192 bits; the second tail bit generator 122 outputs 96 bits; the third tail bit generator 123 outputs 54 bits; and the fourth tail bit generator 124 outputs 30 bits. First to fourth encoders 11-14 encode data output from the first to fourth tail bit generators 121-124, respectively. Here, a convolutional encoder having a constraint length of K=9 and a coding rate of R=l/3 can be used for the encoders 11-14. In this case, the first encoder 11 encodes the 192-bit data output from the first tail bit generator 121 into 576 symbols of full rate; the second encoder 12 encodes the 96-bit data output from the second tail bit generator 122 into 288 symbols of 1/2 rate; the third encoder 13 encodes the 54-bit data output from the third tail bit generator 123 into 162 symbols of about 1/4 rate; and the fourth encoder 14 encodes the 30-bit data output from the fourth tail bit generator 124 into 90 symbols of about 1/8 rate. The rate matcher 20 includes repeaters 22-24 and symbol deletion devices - 4 - 27-28. The repeaters 22-24 repeat symbols output from the second to fourth encoders 12-14 predetermined times, respectively, so as to increase output symbol rates thereof to the full rate. The symbol deletion devices 27 and 28 delete symbols output from the repeaters 23 and 24, which exceed the symbols of the full rate in number. Since the second encoder 12 outputs 288 symbols which is 1/2 the 576 symbols output from the first encoder 11, the second repeater 22 repeats the received 288 symbols two times to output 576 symbols. Further; since the third encoder 13 outputs 162 symbols which is about 1/4 the 576 symbols output from the first encoder 11, the third repeater 23 repeats the received 162 symbols four times to output 648 symbols, which exceeds the 576 symbols of full rate in number. To match the symbol rate to the full rate, the symbol deletion device 27 deletes every ninth symbols to output 576 symbols of full rate. In addition, since the fourth encoder 14 outputs 90 symbols which is about 1/8 the 576 symbols output from the first encoder 11, the fourth repeater 24 repeats the received 90 symbols eight times to output 720 symbols, which exceeds the 576 symbols of full rate in number. To match the symbol rate to the full rate, the symbol deletion device 28 deletes every fifth symbols to output S76 symbols of full rate. First to fourth channel interleavers 31-34 interleave the symbols of full rate output from the first encoder 11„ the second repeater 22, the symbol deletion device 27 and the symbol deletion device 28, respectively. Forward error correction (FEC) is used to maintain a sufficiently low bit error rate (BER) of a mobile station for a channel having a low signal -to-noise ratio (SNR) by providing a channel coding gain. The forward link for the multicarrier communication system can share the same frequency band with the forward link for - 5 - the existing IS-95 system in an overlay method. However, this overlay method raises the follow problems. In the overlay method, three forward link carriers for the multicarrier system are overlaid on three 1.25MHz bands used in the existing IS-95 CDMA system. ?XG. 3 illustrates transmission power levels, by the respective bands, of base stations for the IS-95 system and the multicarrier system. In the overlay method, since the frequency bands for the multicarrier system are overlaid on the frequency bands for the existing IS-95 system, the transmission power or channel capacity is shared between the IS-95 base station and the multicarrier base station at the same frequency band. In the case where the transmission power is shared between the two systems, the transmission power is first allocated for the IS-95 channel which mainly supports a voice service and then, the maximum transmission power permissible to the respective carriers for the multicarrier CDMA system is determined. Here, the maximum transmission power cannot exceed a predetermine power level, because the base station has a limited transmission power. Further, when the base station transmits data too many subscribers, interference among the subscribers increases resulting in an increase in noises. FIG- 3 illustrates the state where the IS-95 base station and the multicarrier base station allocate almost equal transmission powers at the respective 1.25MHz frequency bands. However, the IS-95 channels of 1.2SMHz frequency bands have a different transmission power according to a change in the number of the subscribers in service and a change in voice activity of the subscribers. FIGs. 4 and 5 illustrate the situations where the transmission power allocated for the multicarrier base station decreases at some carriers, as the transmission power allocated for the IS-95 base - 6 - station increases rapidly at the corresponding frequency bands due to an increase in number of the IS-95 subscribers. As a result, sufficient transmission power cannot not be allocated for one or more of the multiple carriers, so that the SNRs are different according to the carriers at the receiver. Accordingly, a signal received at a carrier having the low SNR increases in a bit error rate (BER). That is, when the tS-95 subscriber increases in number and the voice activity is relatively high, a signel transmitted via a carrier overlaid on the corresponding frequency band increases in the BER, resulting in a decreased system capacity and an increased interference among the IS-95 subscribers. That is, the overlay method may cause a reduction in capacity of the multicarrier system and an increase in interference among the IS-95 subscribers. In the multicarrier system, the respective carriers may have independent transmission powers as illustrated in FIGs. 4 and 5. In the light of the performance, FIG. 4 shows the power distribution which is similar to the case where a R-l/2 channel encoder is used, and FIG. 5 shows the power distribution which is worse than the case where the channel encoder is not used. In these cases, one or two of the three coded bits (i.e., symbols) for an input data bit may not be transmitted, causing a degradation of the system performance. Moreover, even in a direct spreading CDMA communication system using a single carrier, weight distribution of the symbols generated by channel encoding is poor, which may cause a performance degradation of channel decoding. SUMMARY OF THE INVENTION - 7 - It is, therefore, an object of the present invention to provide a channel encoding device and method capable of generating coded data having a good channel coding performance in a CDMA communication system. It is another object of the present invention to provide a channel encoding device and method capable of generating channel coded data having a good channel coding performance and effectively distributing the generated channel-coded data to respective earners in a multicarrier CDMA communication system. It is further another object of the present invention to provide a channel transmission device and method for distributing generated symbols to carriers such that an influence of symbols damaged during transmission can be minimized in a multicarrier CDMA communication system. It is still another object of the present invention to provide an R=l/6 convolutional encoding device and method capable of increasing a channel performance in a channel transmitter for a CDMA communication system. To achieve the above object, there is provided a communication system using at least two carriers. The communication system includes a channel encoder for encoding data, a channel controller for generating a control signal for transmitting channel coded symbols such that decoding can be performed using data received via at least one carrier, and a symbol distributer for assigning the channel coded symbols to at least two carriers. BRIEF DESCRIPTION OF THE DRAWINGS - 8 - The above and other objects, features and advantages of the present invention will become more apparent from the following detailed description when taken in conjunction with the accompanying drawings, in which: FIG. 1 is a diagram illustrating & forward link structure for a conventional multicarrier CDMA communication system; FIG. 2 is a diagram illustrating & fundamental channel structure for a forward link of FIG. 1; FIG. 3 is a diagram illustrating transmission power distribution of IS-95 channel bands and multicarrier channel bands in the case where the multicarrier channels are overlaid on the 1S-95 channels at the same frequency bands; FIG. 4 is a diagram illustrating a state where the transmission power for one of multiple carriers is decreased when a transmission power for a corresponding IS-95 channel is increased, due to a limitation in transmission power or transmission capacity of the system; FIG. 5 is a diagram illustrating a state where the transmission powers for two of multiple carriers are decreased when transmission powers for corresponding 1S-95 channels are increased, due to a limitation in transmission power or transmission capacity of the system; FIG. 6 is a diagram illustrating a scheme for generating convolutional codes of a symbol rate 1/6, employing a channel encoder and a symbol distributer according to an embodiment of the present invention; FIG 7 is a detailed diagram illustrating an R=l/6 convolutional encoder of FIG. 6; FIG. 8 is a detailed diagram illustrating a symbol distributer of FIG. 6, FIG. 9 is a diagram illustrating a transmission scheme for a forward link - 9 - using a channel encoder and a symbol distributer according to an embodiment of the present invention, FIG. 10 is a simulation diagram illustrating a performance comparison among R=l/3 convolutional codes according to an embodiment of the present invention; FIG. 11 is a simulation diagram illustrating a worst performance comparison among R= 1 /2 convolutional codes using generator polynomials of a convolutional encoder having a coding rate of R= 1/3; FIG. 12 is a simulation diagram illustrating a performance comparison among R=l/2 constraint codes for an R=V6 convolutional code; and FIG. 13 is a simulation diagram illustrating a worst performance comparison among R=l/2 constraint codes using an R=1/6 convolutional encoder with a highest performance. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT A preferred embodiment of the present invention will be described hereinbelow with reference to the accompanying drawings. In the following description, well known functions or constructions are not described in detail since they would obscure the invention in unnecessary detail. A term "symbol" as used herein refers to a coded data bit output from an encoder. For convenience of explanation, it is assumed that the multicarrier communication system is a three-carrier CDMA communication system using three carriers. - 10 - In a communication system supporting both the IS-95 system and the multicarrier system, wherein transmission signals of the two different systems are overlaid at the same frequence bands, coded symbols are distributed such that a performance degradation may be minimized during decoding of the damaged symbols, and then the distributed coded bits are assigned to the respective carriers. Thus, even if one of the carriers has interference during reception, it is possible to perform decoding for only the coded bits transmitted via the other carriers, thereby improving the system performance. Moreover, in the forward link, an R=l/6 convolutional code can be used for a channel encoder. Therefore, when the channel encoder generates R=1/6 convolutional codes, it is very difficult to find R=l/6 convolutional codes having a good decoding performance. Accordingly, the present invention is directed to generating R=l/6 convolutional codes with a good decoding performance and distributing the generated convolutional codes to multiple carriers. The R=/16 convolutional codes generated according to the present invention have a good performance in both a multicamer CDMA communication system and a DS-CDMA communication system, A description will now be made regarding an operation of generating symbols for maximizing a channel performance and distributing the generated symbols in a CDMA communication system according to an embodiment of the present invention. For convenience, the present invention will be described hereinafter with reference to a multicarrier CDMA communication system First, reference will be made to R-l/6 convolutional codes for a multicamer - 11 - CDMA communication system using three carriers. FIG. 6 illustrates a convolunonal encoder and a symbol distributer according to an embodiment of the present invention. Referring to FIG. 6, a convolutional encoder 601 encodes one input data bit into six symbols which are allocated to three carriers A, B and C. For symbol allocation, a symbol distributer 602 uniformly distributes the six input bits to the three carriers by two bits. The symbol distributer 602 distributes the symbols output from the convolutional encoder 601, taking into consideration how many carriers are damaged out of the three carriers. By using this symbol distribution method, even though one or two out of the three carriers are damaged, performance degradation in channel decoding can be minimized. A description will now be made as to a method of designing the symbol distributer 602. A bit error rate (BER) after channel decoding depends on a damaged portion for the symbols coded by a channel encoder. Therefore, even though the coded symbols are damaged, the symbols located at the minimized performance degraded portion are uniformly distributed to the carriers. Accordingly, even though the symbols for a certain channel arc all damaged, an increase in the BER after channel decoding can be minimized. In addition, during transmission, the symbols output from a constituent encoder in the channel encoder are distributed to the carriers; during decoding, a constituent decoder in a channel decoder is selected such that the BER can be low even though the symbols for a certain carrier are all damaged. - 12 - Selection of the constituent decoder in the channel decoder is made in the following process. First, reference will be made to a convolutional code having a constraint length of K=9 and a rate of R=l/3. In the following descriptions, generator polynomials gi are represented by octal number. The convolutional code with K=9 and R= 1/3 has a free distance of dfm= 18. It is noted that there exists 5685 sett, when search is made for convolutional codes having K=9, R-l/3 and dfree 18, by changing generator polynomials g1, g2 and g3. Here, only non-catastrophic codes are selected. In addition, it is necessary to prevent the performance degradation even though a certain carrier is completely off, providing for application to the multicarrier system. From this point of view, it is preferable to maximize the free distance. For a reference code for performance comparison, a convolutional code of (g1, g2 ,g3)= (557, 663, 711) is used which is used in the existing IS-95 system. In the IS-95 system, a free distance of the convolutional code is dfree=18, and free distances between constituent codes are dfree(g3S7,g661)=9, dfreer(g557,g711)=11,and dfree (g663,.g711)=10. A performance of a convolutional code can be predicted using a BER upper limit formula, which is determined by a transfer function. For the IS-95 system, a transfer function of a convolutional code is T(D,1), = 5D18+7D20+O(D21), and a BER upper limit formula is 1 lD18+32D26+O(D21). When the convolutional code for the IS-95 system is view in the light of a constituent code, a catastrophic error propagation occurs at a combination of generator polynomials g, and g2. Therefore, when the convolutional codes for the IS-95 system are used for the multicarrier system, it is necessary to appropriately use interleaving and puncturing. Since the IS-95 convolutional codes - 13 - have the catastrophic error propagation in the light of the constituent codes, it is necessary to search for new convolutional codes suitable for the multicarrier system. For K=9, dfree(g1,g2)11. Here, not only the codes but also the constituent codes are non-catastrophic Since a first term of the BER upper limit formula is most influential, first and eighth codes can be considered to be most optimal codes. Here, since the pairs of first and eighth codes; second and seventh codes; third and fourth codes; and fifth and sixth codes are in reciprocal relation, they are the same codes essentially Therefore, there exist only four codes. Table 1 is given to explain a characteristic of a convolutional encoder with K=9 and R=l/3. [TABLE 1] - 14 - , 15 - In Table 1, dI2 in a first term means d(447,543) and hereinafter, used in the same meaning. For information, when the codes are compared with the IS-95 codes in the light of the first term of the BER upper limit formula, the first and eighth codes are superior in performance to the IS-95 codes, the third, fourth, fifth and sixth codes are similar in performance to the IS-95 codes, and the second and seventh codes are inferior in performance to the IS-95 codes Therefore, it is preferable to use the eighth (or first) code. In the meantime, there exist four or more codes whose the free distances among the constituent codes are 12, 12 and 10; among these codes, a generator polynomial for a superior code in the light of the first term of the BER upper limit formula is (g1,g2,g3) = (515, 567, 677). Shown in FIG. 10 is the simulation Tesult for performances of the convolutional code with R=l/3 in a AWGN (Additive White Gaussian Noise) environment in the case where the multicarrier (three-carrier) system has an optimal performance without damage of the respective carriers. In the following descriptions, the simulations of FIGs. 11-13 are all performed in the AWGN environment. represents a R=l/3 convolutional code for an existing IS-95 system, and and represent a R=l/3 - 16 - convolutionsl code searched for in the above method. A description will now be made regarding the case where the R=l/3 convolutional code is applied to the three-carrier system and a certain one of the three carriers is damaged (or lost). Although the original coding rate is 1/3, the loss of one carrier causes the coding rate to be equal to 1/2. Therefore, shown in FIG 11 is the simulation results for the 1/2 convolutional codes using the generator polynomials for the 1/3 convolutional codes. In FIG. 11, the respective conditions can be explained by the following through . FIG. 11 illustrates the worst performance graph for the R=l/2 convolutional codes using the generator polynomials for the R=l/3 convolutional code. Optimal 1/2 convolutional code - g,=56l, g2=753, dfree(g1.g2)=12 the worst performance, g,=557, g2=711 out of three R=l/2 convolutional codes using the generator polynomial (557, 663, 711) for a 1/3 convolutional code used for the IS-95 system - catastrophic error propagation occurs the worst performance, g,=731, g,=615 (dfree,(gl,g2)=11) for a R=l/2 convolutional code using the generator polynomial (731, 615, 537) for a R=l/3 convolutional code - 17 - the worst perforrnance, g,=567, g;=677 (dfree(gl,g1)=10) for a R=l/2 convolutional code using the generator polynomial (515, 567, 677) for a R=l/3 convolutional code When one carrier is damaged in a three-carrier system using an R=l/3 convolutional code, the coding rate becomes to be equal to R=l/2. In this case, a symbol distribution method for the symbol distributer is found by appropriately distributing the original R=l/3 convolutional codes to the three carriers using the following symbol deleting matrixes, so as to minimize the performance degradation even though the coding rate becomes R=l/2. In the simplest method, the following two symbol deleting matrixes are generated. In the following symbol deleting matrixes, "0" means the case where a carrier to which the corresponding symbol is provided is damaged, and "I" means the case where the carrier to which the corresponding symbol is provided is not damaged. That is, this means the case that the symbols corresponding to "0" are all allocated to a certain carrier, which is damaged during transmission. Therefore, one of the following various patterns of the symbol deleting matrix is selected, which minimizes the performance degradation even though one carrier is damaged, and the symbol distributer 602 provides the symbols to the respective carriers using the selected pattern. The followings are symbol deleting matrixes for finding a pattern used for the symbol distributer 602. - 18 - Further an m-sequence of length=8 is generated over a two-stage GF(3) using an m-sequence. For a ninth convolutional code, a sequence (1,2,0,2,2,1,0,1,2) is generated and then, the following symbol deleting matrix Ds is generated using the sequence. Further, the following symbol deleting matrixes D4 and D5 are generated by * changing the row of the symbol deleting matrix D3 In addition, a sequence {2,1,0,1,1,0,1.2,1,0,0,0,2,1,2} is obtained by generating 15 random numbers over GF(3) using a random number, and the - 19 - following symbol deleting matrix D4 is created using the above sequence Also, the following symbol deleting matrixes D7 and D8 are generated by changing the rows as in the method using the m-sequence Next, a description will be made regarding a convolutional code having a symbol rate of 1/6. A K=9, R=l/6 convolutional code has a free distance of dfree=37. In searching for the convolutional codes having a free distance of dfree=37 by randomly changing the generator polynomials g1 g2 ..., g3. the following conditions should be satisfied First, it should be a R=l/6 convolutional code with a good decoding performance. Second, it should be a R=l/4 convolutional code with a good decoding - 20 - performance which has generator polynomials (g1,g2,g3,g4), (g1,g2,g5,g6) and (g1,g2,g5,g6), considering the case that one of three carriers is damaged in the three-carrier system. Third, it should be a 1/2 convolutional code with a good decoding performance which has generator polynomials (g1,g2), (g3.g4 and (g5,g6), considering the case that two of three carriers are damaged in the three-carrier system. In the second and third conditions out of the above three conditions, the performance degradation is minimized even though one or two of the three carriers are completely off, providing for the multicarrier system in which six output bits of the covolutional code are allocated to three carriers by two bits. From this point of view, it is preferable that the R=l/4 convolutional code and the R=1/2 convolutional code have the maximum free distance. A method of searching for a R=l/2 convolutional code satisfying the third condition becomes apparent from the following description. There exist 35 non-catastrophic convolutional codes with R=l/2, K=9 and dfree=12. An upper limit formula for the BER is give as follows, and a coefficient c12 of the most important term D12 in determining the BER ranges from 33 to 123. First, for the R=l/6 convolutional codes, there exist 180 R=l/6 convolutional codes with dfree=37, satisfying the third condition. It is assumed that - 21 - dfree(g1,g2,g5,g6)=12. Here, there exist 58 convolutional codes in which the first term of the BER upper limit formula for the R=l/6 convolutional code has a coefficient of C37=l. The following are the R=l/6 convolutional codes selected among the 58 convolutional codes after performance verification. - 22 - The following are 5 R=1/6 convolutional codes with a good decoding performance selected among the 12 performance-verified l/6 convolutional codes. A performance of the R-1/2 convolunonal codes using five generator polynomials for the R=1/6 convolutional code are verified, and further, a performance of the R= 1/4 convolutional codes using five generator polynomials for the R-1/6 convolutional code are verified. First, a transfer function for the R=l/2 convolutional codes will be described with reference to Table 2 in which the generator polynomials are represented by octal number. [TABLE 2] No 1/2 Generator Polynomial BER Upper Limit Formula, Coeff c12, c12, c14 1 435,657 (33,0,281) 2 561,753 (33,0,281) 3 515,677 (38, 106,238) 4 545, 773 (38, 106,238) 5 463, 755 (38,0,274) 6 557,631 (38,0,274) 7 557,751 (40,33, 196) 8 .457,755 (40,33, 196) 9 453,755 (40,0,271) 10 :557, 651 (40,0,271) 11 471,673 (50,0,298) 12 537,615 (50,0,360) - 23 - - 24 - A R=1/2 convolutional code with the highest performance is searched for by verifying the perfonnances of the respective R=1/2 convolutional codes in Table 2. In addition, performances of the R=l/2 convolutional codes are compared with performances of the optimal R=l/2 convolutional code used for the IS-95 system. A performance comparison among the respective cases is shown in FIG. 12 FIG. 12 illustrates a performance comparison among R=l/2 constituent codes for the R=l/6 convolutional code. It is noted that the R=l/2 constituent codes for the R=l/6 convolutional code arc similar in performance to the optimal R=l/2 - 25 - convolutional code. Table 3 illustrates transfer functions for the R=1/6 convolimonal codes. - 26 - [TABLE 3] - 27 - - 2B ¦ The worst performances of the R=l/2 constituent codes using 5 R=1/6 cnnvolutional codes with the good decoding performance are as follows, with reference to Table 3. the worst performance of a R=l/6 convolutional code (NO=1) having generator polynomials of (457, 755, 551, 637, 523, 727), - (523, 727)g, cl2 = 68 the worst performance of a R=l/6 convolutional code (NO-7) having generator polynomials of (515, 677,453, 755, 551, 717), - (515,677),. cl2 = 38 the worst performance of a R=l/6 convolutional code (NO=8) having generator polynomials of (545, 773, 557, 651, 455, 747)g - (545, 773),, cl2 = 38 the worst performance of a R=l/6 convolutional code (NO=11) having generator polynomials of (551, 677, 557, 651, 455, 747)R - (551, 677)g. c12 , - 29 - = 38 the worst performance of a R=l/6 convolutional code (NO=29) having generator polynomials of (515, 677, 455,747,531,657), - (515, 677),, c12 = 38 The wont performances for the R= 1/4 constituent codes are as follows using the R=1/6 convolutional codes whose performances are verified for the R=l/2 constituent codes. the worst performance of s R=1/6 convoltional code (NO-1) having generator polynomials Of (457, 755, 551,637,523, 727), - (551,637, 523, 727)g, c24 = 5 the worst performance of a R-l/6 convolutional code (NO=7) having generator polynomials of (515, 677,453, 755, 551, 717), - (515,677, 551 the worst performance of a R=)/6 convoiutional code (NO=8) having generator polynomials of (545, 773, 557,651, 455, 747), - (545, 773, 455, the worst performance of a R=l/6 convolutional code the worst performance of a R=l/6 convolutional code (NO=29) having generator polynomials of (515, 677,455,747, 531,657), - (5 15, 677, 531, FIG, 13 iilustrates a comparison among the worst performances of R= F/2 - 30 - constituent codes using R=l/6 convolutional code with the highest performance. The following are two R=l/6 convolutional codes with the good decoding performance, selected among the R=l/6 convolutional codes whose performances are verified for various cases in the above manners. - 31 - Further, to search for a symbol deleting pattern used for the three-carrier system, various symbol deleting matrixes are considered for the situation where one carrier is damaged, i.e., where the R=l/6 convolutional codes change to R=l/4 convolutional codes. The reason for searching for the symbol deleting matrix pattern is the same as described for the R=l/3 convolutional codes. The following matrixes can be used as a symbol deleting matrix pattern for a method of distributing symbols for R=l/6 convolutional codes. -32- - 33 - Taking into consideration the case where two carriers are damaged in the three-carrier system, the following symbol deleting matrix pattern can be used in a method of distributing symbols for R=1/2 symbol-deleted convolutional codes using generator polynomials for the R=l/6 convolutional codes with a good decoding performance. Turning to FIG. 6, there is shown the convolutional encoder 601 and the - 34 - symbol distributer 602 according to an embodiment of the present invention. In the exemplary embodiment, the convolutional encoder 601 has a coding rate of R=l/6 and uses generator polynomials of (545, 773, 557, 651, 455, 747). The detailed structure of the R=l/6 convolutional is illustrated in FIG. 7. Referring to FTG. 7, upon receipt of input data, delays 711 -A to 711-H delay the input data bits sequentially. During the sequential delay of the input data bits, exclusive OR gates 721-A to 721-F output coded symbols. The coded symbols of FIG. 7 are provided to the symbol distributer 602 having the structure of FIG. 8 Referring to FIG. 8, the symbol distributer 602 is implemented by switches 811-A and 811-B. In FIG 8, when a symbol rate of a clock for controlling the switches 811-A and 811-B is over six times a symbol rate of the symbol distributer 602, the symbols can be distributed without symbol loss. That is, the switch 811-A sequentially receives input symbols g1, g2, g3, g4, g5, g6, g1, g2, g3 and the switch 811-R distributes the input symbols to output nodes C1, c2, c3, c4 c5 and c6 FIG. 9 illustrates a transmission scheme including the channel encoder 601 and the symbol distributer 602 of FIG. 6 Referring to FIG. 9, first to fourth CRC generators 911-914 add CRC data in a specified number of bits to input data. Specifically, 12-bit CRC is added to the 172-bit data of the first rate; 8-bit CRC is added to the 80-bit data of the second rate; 6-bit CRC is added to the 40-bit data of the third rate; and 6-bit CRC is added to the 16-bit data of the fourth rate. First to fourth tail bit generators 921-924 add 8 tail bits to the CRC-added data. Therefore, the first tail bit generator 921 outputs - 35 - 192 bits; the second tail bit generator 922 outputs 96 bits; the third tail bit generator 923 outputs 54 bits; and the fourth tail bit generator 924 outputs 30 bits. First to fourth encoders 931 -934 encode data output from the first to fourth tail bit generators 921-924, respectively. Here, a K=9, R=l/6 convolutional encoder can be used for the encoders 931-934. In this case, the first encoder 931 encodes the 192-bit data output from the first tail bit generator 921 into 1152 symbols of full rate; the second encoder 932 encodes the 96-bit data output from the second tail bit generator 922 into 576 symbols of 1/2 rate; the third encoder 933 encodes the 54-bit data output from the third tail bit generator 923 into 324 symbols of about 1/4 rate; and the fourth encoder 934 encodes the 30-bit data output from the fourth tail bit generator 924 into 180 symbols of about 1/8 rate. First to fourth symbol distributers 941-944 distribute the symbols output from the encoders 931-934, respectively. Here, for symbol distribution, a channel controller (not shown) generates control signals for distributing the channel coded bits such that the performance degradation may be minimized during decoding of received damaged bits, when the coded symbols are transmitted being overlaid on the symbols of a different system at the same frequency band. The symbol distributers 941-944 then assign the symbols output from the encoders 931-934 to the corresponding carriers according to the control signals, respectively. Rate matchers 951 -953 each include a symbol repeater and a symbol deletion device. The rate matchers 951-953 match rates of the symbols output from the corresponding symbol distributers 942-944 to a rate of the symbols output from the symbol distributer 941. First to fourth channel interleavers 961-964 interleave the - 36 - symbols output from the symbol distributer 941 and the rate matchers 951-953, respectively. For the DS-CDMA communication system, the symbol distributers 941-944 of FIG, 9 can be removed. As described above, in the multicarrier system employing the frequency overlay method, the respective carriers have limited transmission powers according to the loading in the frequency bands of the existing 1S-95 system, which results in loss of data received at one or more carrier frequency bands. To solve this problem, by using the generator polynomials for the channel encoder and a symbol distribution method, it is possible to provide a high coding gain against the data loss due to the carrier loss, thereby preventing degradation of the BER. While the invention has been shown and described with reference to a certain preferred embodiment thereof, it will be understood by those skilled in the art that various changes in forrn and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims. - 37 - -38- We Claim 1. A channel transmission device for a multi-carrier CDMA communication system using at least two carriers, comprising: a channel encoder (601, 931-934) for encoding channel data to be transmitted into symbols at a predetermined coding rate; and a symbol distributor (602, 941-944) for receiving the symbols and distributing received symbols to the carriers according to a predetermined symbol deleting matrix pattern, wherein the symbol deleting matrix pattern is so determined as to distribute the symbols to the respective carriers with a minimized performance degradation even though a specific carrier is damaged. 2. The channel transmission device as claimed in claim 1, further comprising: a channel interieaver (961-964) for channel interleaving the distributed symbols; a demultiplexer (93) for distributing the interleaved symbols to the carriers, a plurality of orthogonal modulators (61-63) for generating orthogonally modulated signals by multiplying the distributed symbols by orthogonal codes for the corresponding channels; a plurality of spreaders (71-73) for receiving the orthogonally modulated signals and generating spread signals by multiplying the received orthogonally modulated signals by a spreading code; and a plurality of transmitters (81-83) for receiving the spread signals and transmitting the received spread signals using the carriers. -39- 3. The channel transmission device as claimed in claim 1 or 2, wherein the channel encoder is a convolutional encoder with a coding rate of R=l/6. 4. The channel transmission device as claimed in one of claims 1 to 3, wherein the symbol distributor comprises: a first selector (811-A) for sequentially multiplexing the received symbols; and a second selector (811-B) for distributing the multiplexed symbols to the carriers according to the symbol distribution signal. 5. The channel transmission device as claimed in one of claims 1 to 4, further comprising: a channel controller for generating a symbol distribution signal according to the predetermined symbol deleting matrix pattern, wherein the symbol distributor is arranged for distributing the received symbols to the carriers according to the symbol distribution signal. 6. A channel transmission method for a CDMA communication system using at least two carriers, comprising the steps of: encoding channel data to be transmitted into symbols at a predetermined coding rate; and receiving the symbols and distributing received symbols to the carriers according to a predetermined symbol deleting matrix pattern, wherein the symbol deleting matrix pattern is so determined as to distribute the symbols to the respective carriers with a minimized performance degradation even though a specific carrier is damaged. -40- 7. The channel transmission method as claimed in claim 6, wherein the channel data is encoded using a convolutional encoder with a coding rate of R=1/6. 8. The channel transmission method as claimed in claim 6 or 7, wherein the symbol distribution step comprises the step of: sequentially multiplexing the received symbols; and distributing the multiplexed symbols to the carriers according to the symbol distribution signal. 9. The channel transmission device as claimed in claim 1, wherein the channel encoding device comprising; a plurality of delays for delaying an input data bit to generate first to eight delayed data bits; a first operator for exclusively ORing the input data bit and the third, fifth, sixth, seventh and eighth delayed data bits to generate a first symbol; a second operator for exclusively ORing the input data bit and the first, second, third, fifth, sixth and eighth delayed data bits to generate a second symbol; a third operator for exclusively ORing the input data bit and the second, third, fifth and eighth delayed data bits to generate a third symbol; a fourth operator for exclusively ORing the input data bit and the first, fourth, fifth, sixth, seventh and eighth delayed data bits to generate a fourth symbol; -41- a fifth operator for exclusively ORing the input data bit and the first, fourth, sixth and eighth delayed data bits to generate a fifth symbol; and a sixth operator for exclusively ORing the input data bit and the first, second, fourth, sixth, seventh and eighth delayed data bits to generate a sixth symbol. 10. The channel transmission method as claimed in claim 6, wherein the channel encoding method for a CDMA communication system, comprising the steps of: shiftingly delaying an input data bit to generate first to eight delayed data bits; exclusively ORing the input data bit and the third, fifth, sixth, seventh and eighth delayed data bits to generate a first symbol; exclusively ORing the input data bit and the first, second, third, fifth, sixth and eighth delayed data bits to generate a second symbol; exclusively ORing the input data bit and the second, third, fifth and eighth delayed data bits to generate a third symbol; exclusively ORing the input data bit and the first, fourth, fifth, sixth, seventh and eighth delayed data bits to generate a fourth symbol; exclusively ORing the input data bit and the first, fourth, sixth, and eighth delayed data bits to generate a fifth symbol; and exclusively ORing the input data bit and the first, second, fourth, sixth, seventh and eighth delayed data bits to generate a sixth symbol. A channel transmission device for a code division multiple access (CDMA) communication system using at least two carriers. In the channel transmission device, a channel encoder encodes channel data to be transmitted into symbols at a predetermined coding rate. A channel controller generates a symbol distribution signal according to a predetermined symbol deleting matrix pattern. The symbol deleting matrix pattern is so determined as to distribute the symbols to the respective carriers with a minimized performance degradation even though a specific carrier is damaged. A symbol distributer receives the symbols and distributes received symbols to the carriers according to the symbol distribution signal.

Full Text

PRIORITY
This application claims priority to an application entitled "Device and Method for Generating and Distributing Coded Symbols in CDMA Communication System" filed in the Korean Industrial Property Office on May 30, 1998 and assigned Serial No. 98-2014U,
BACKGROUND OF THE INVENTION
1, Field of the Invention
The present invention relates generally to a data transmission device and method for a CDMA communication system, and in particular, to a device and method for generating and distributing symbols capable of preventing degradation of a channel performance during data transmission.
2. Description of the Related Art
At present, code division multiple access (CDMA) communication systems are implemented based on the IS-95 standard. However, with the development of communication technology, subscribers to the communication services increase greatly in number. Therefore, there are proposed many methods for meeting the subscriber's increasing demands for the high quality service. An approach to such

methods includes a method for improving a forward link structure.
For an improved forward link structure, there is a forward link fundamental channel designed for a third generation multicarrier CDMA system proposed in the T1A/E1A TR45,5 conference A forward link structure for a multicarrier CDMA communication system is illustrated in FIG. 1.
Referring to FIG. 1, a channel encoder 10 encodes input data, and a rate matcher 20 repeats and punctures symbols output from the channel encoder 10. Here, the data input to the channel encoder 10 has a variable bit rate The rate matcher 20 repeats and punctures the coded data bits (i.e., symbols) output from the channel encoder 10 so as to match symbol rates for the data having the variable bit rate. A channel interleaver 30 interleaves an output of the rate matcher 20. A block interleaver is typically used for the interleaver 30,
A long code generator 91 generates a long code which is identical to that used by the subscriber. The long code is a unique identification code for the subscriber. Thus, different long codes are assigned to the respective subscribers. A decimator 92 decimates the long code to match a rate of the long code to a rate of the symbols output from the interleaver 30. An adder 93 adds an output of the channel interleaver 30 and an output of the decimator 92 An exclusive OR gate is typically used for the adder 93
A demultiplexer 40 sequentially multiplexes data output from the adder 93 to multiple carriers A, B and C. First to third binary-to-four level converters 51-53 convert signal levels of binary data output from the demultiplexer 40 by converting
- 2 -

input data of "0" to "+l" and input data of "V to "-\". First to third orthogonal modulators 61-63 encode data output from the first to third level converts 51-53 with corresponding Walsh codes, respectively. Here, the Walsh codes have a length of 256 bits. First to third spreaders 71-73 spread outputs of the first to third orthogonal modulators 61-63, respectively. Here, QPSK. (Quadrature Phase Shift Keying) spreaders can be used for the spreaders 71-73. First to third attenuators (or gain controller) 81-83 control gains of the spread signals output from the first to third spreaders 71-73 according to corresponding attenuation signals GA-GC, respectively. Here, the signals output from the first to third attenuators 81-83 become different earners A, B and C.
In the forward link structure of FIG. 1, the channel encoder 10 having a coding rate of R=l /3 encodes the input data into 3 coded data bits (i.e., code words or symbols) per bit. Such coded data bits are demultiplexed to the three carriers A, B and C after rate matching and channel interleaving.
The multicarrier CDMA communication system of FIG. 1 can be modified to a single carrier CDMA communication system by removing the demultiplexer 40 and using only a set of the level converter, the orthogonal modulator, the spreader and the attenuator.
FIG. 2 is a detailed diagram illustrating the channel encoder 10, the rate matcher 20 and the channel interleaver 30. In FIG. 2, data of a first rate is composed of 172 bits (full rate) per 20ms frame; data of a second rate is composed of 80 bits (1/2 rate) per 20ms frame; data of a third rate is composed of 40 bits (1/4 rate) per 20ms frame; and data of a fourth rate is composed of 16 bits (1/8 rate) per
- 3 -

Referring to FIG. 2, first to fourth CRC generators 111-114 generate CRC bits corresponding to the respective input data having different rates and add the generated CRC bits to the input data. Specifically, 12-bit CRC is added to the 172-bit data of the first rate; 8-bit CRC is added to the 80-bit data of the second rate; 6-bit CRC is added to the 40-bit data of the third rate; and 6-bit CRC is added to the 16-bit data of the fourth rate.
First to fourth tail bit generators 121-124 add 8 tail bits to the CRC-added data, respectively. Therefore, the first tail bit generator 121 outputs 192 bits; the second tail bit generator 122 outputs 96 bits; the third tail bit generator 123 outputs 54 bits; and the fourth tail bit generator 124 outputs 30 bits.
First to fourth encoders 11-14 encode data output from the first to fourth tail bit generators 121-124, respectively. Here, a convolutional encoder having a constraint length of K=9 and a coding rate of R=l/3 can be used for the encoders 11-14. In this case, the first encoder 11 encodes the 192-bit data output from the first tail bit generator 121 into 576 symbols of full rate; the second encoder 12 encodes the 96-bit data output from the second tail bit generator 122 into 288 symbols of 1/2 rate; the third encoder 13 encodes the 54-bit data output from the third tail bit generator 123 into 162 symbols of about 1/4 rate; and the fourth encoder 14 encodes the 30-bit data output from the fourth tail bit generator 124 into 90 symbols of about 1/8 rate.
The rate matcher 20 includes repeaters 22-24 and symbol deletion devices
- 4 -

27-28. The repeaters 22-24 repeat symbols output from the second to fourth encoders 12-14 predetermined times, respectively, so as to increase output symbol rates thereof to the full rate. The symbol deletion devices 27 and 28 delete symbols output from the repeaters 23 and 24, which exceed the symbols of the full rate in number. Since the second encoder 12 outputs 288 symbols which is 1/2 the 576 symbols output from the first encoder 11, the second repeater 22 repeats the received 288 symbols two times to output 576 symbols. Further; since the third encoder 13 outputs 162 symbols which is about 1/4 the 576 symbols output from the first encoder 11, the third repeater 23 repeats the received 162 symbols four times to output 648 symbols, which exceeds the 576 symbols of full rate in number. To match the symbol rate to the full rate, the symbol deletion device 27 deletes every ninth symbols to output 576 symbols of full rate. In addition, since the fourth encoder 14 outputs 90 symbols which is about 1/8 the 576 symbols output from the first encoder 11, the fourth repeater 24 repeats the received 90 symbols eight times to output 720 symbols, which exceeds the 576 symbols of full rate in number. To match the symbol rate to the full rate, the symbol deletion device 28 deletes every fifth symbols to output S76 symbols of full rate.
First to fourth channel interleavers 31-34 interleave the symbols of full rate output from the first encoder 11„ the second repeater 22, the symbol deletion device 27 and the symbol deletion device 28, respectively.
Forward error correction (FEC) is used to maintain a sufficiently low bit error rate (BER) of a mobile station for a channel having a low signal -to-noise ratio (SNR) by providing a channel coding gain. The forward link for the multicarrier communication system can share the same frequency band with the forward link for
- 5 -

the existing IS-95 system in an overlay method. However, this overlay method raises the follow problems.
In the overlay method, three forward link carriers for the multicarrier system are overlaid on three 1.25MHz bands used in the existing IS-95 CDMA system. ?XG. 3 illustrates transmission power levels, by the respective bands, of base stations for the IS-95 system and the multicarrier system. In the overlay method, since the frequency bands for the multicarrier system are overlaid on the frequency bands for the existing IS-95 system, the transmission power or channel capacity is shared between the IS-95 base station and the multicarrier base station at the same frequency band. In the case where the transmission power is shared between the two systems, the transmission power is first allocated for the IS-95 channel which mainly supports a voice service and then, the maximum transmission power permissible to the respective carriers for the multicarrier CDMA system is determined. Here, the maximum transmission power cannot exceed a predetermine power level, because the base station has a limited transmission power. Further, when the base station transmits data too many subscribers, interference among the subscribers increases resulting in an increase in noises. FIG- 3 illustrates the state where the IS-95 base station and the multicarrier base station allocate almost equal transmission powers at the respective 1.25MHz frequency bands.
However, the IS-95 channels of 1.2SMHz frequency bands have a different transmission power according to a change in the number of the subscribers in service and a change in voice activity of the subscribers. FIGs. 4 and 5 illustrate the situations where the transmission power allocated for the multicarrier base station decreases at some carriers, as the transmission power allocated for the IS-95 base
- 6 -

station increases rapidly at the corresponding frequency bands due to an increase in number of the IS-95 subscribers. As a result, sufficient transmission power cannot not be allocated for one or more of the multiple carriers, so that the SNRs are different according to the carriers at the receiver. Accordingly, a signal received at a carrier having the low SNR increases in a bit error rate (BER). That is, when the tS-95 subscriber increases in number and the voice activity is relatively high, a signel transmitted via a carrier overlaid on the corresponding frequency band increases in the BER, resulting in a decreased system capacity and an increased interference among the IS-95 subscribers. That is, the overlay method may cause a reduction in capacity of the multicarrier system and an increase in interference among the IS-95 subscribers.
In the multicarrier system, the respective carriers may have independent transmission powers as illustrated in FIGs. 4 and 5. In the light of the performance, FIG. 4 shows the power distribution which is similar to the case where a R-l/2 channel encoder is used, and FIG. 5 shows the power distribution which is worse than the case where the channel encoder is not used. In these cases, one or two of the three coded bits (i.e., symbols) for an input data bit may not be transmitted, causing a degradation of the system performance.
Moreover, even in a direct spreading CDMA communication system using a single carrier, weight distribution of the symbols generated by channel encoding is poor, which may cause a performance degradation of channel decoding.
SUMMARY OF THE INVENTION
- 7 -

It is, therefore, an object of the present invention to provide a channel encoding device and method capable of generating coded data having a good channel coding performance in a CDMA communication system.
It is another object of the present invention to provide a channel encoding device and method capable of generating channel coded data having a good channel coding performance and effectively distributing the generated channel-coded data to respective earners in a multicarrier CDMA communication system.
It is further another object of the present invention to provide a channel transmission device and method for distributing generated symbols to carriers such that an influence of symbols damaged during transmission can be minimized in a multicarrier CDMA communication system.
It is still another object of the present invention to provide an R=l/6 convolutional encoding device and method capable of increasing a channel performance in a channel transmitter for a CDMA communication system.
To achieve the above object, there is provided a communication system
using at least two carriers. The communication system includes a channel encoder
for encoding data, a channel controller for generating a control signal for
transmitting channel coded symbols such that decoding can be performed using
data received via at least one carrier, and a symbol distributer for assigning the
channel coded symbols to at least two carriers.
BRIEF DESCRIPTION OF THE DRAWINGS
- 8 -

The above and other objects, features and advantages of the present invention will become more apparent from the following detailed description when taken in conjunction with the accompanying drawings, in which:
FIG. 1 is a diagram illustrating & forward link structure for a conventional multicarrier CDMA communication system;
FIG. 2 is a diagram illustrating & fundamental channel structure for a forward link of FIG. 1;
FIG. 3 is a diagram illustrating transmission power distribution of IS-95 channel bands and multicarrier channel bands in the case where the multicarrier channels are overlaid on the 1S-95 channels at the same frequency bands;
FIG. 4 is a diagram illustrating a state where the transmission power for one of multiple carriers is decreased when a transmission power for a corresponding IS-95 channel is increased, due to a limitation in transmission power or transmission capacity of the system;
FIG. 5 is a diagram illustrating a state where the transmission powers for two of multiple carriers are decreased when transmission powers for corresponding 1S-95 channels are increased, due to a limitation in transmission power or transmission capacity of the system;
FIG. 6 is a diagram illustrating a scheme for generating convolutional codes of a symbol rate 1/6, employing a channel encoder and a symbol distributer according to an embodiment of the present invention;
FIG 7 is a detailed diagram illustrating an R=l/6 convolutional encoder of FIG. 6;
FIG. 8 is a detailed diagram illustrating a symbol distributer of FIG. 6,
FIG. 9 is a diagram illustrating a transmission scheme for a forward link
- 9 -

using a channel encoder and a symbol distributer according to an embodiment of the present invention,
FIG. 10 is a simulation diagram illustrating a performance comparison among R=l/3 convolutional codes according to an embodiment of the present invention;
FIG. 11 is a simulation diagram illustrating a worst performance comparison among R= 1 /2 convolutional codes using generator polynomials of a convolutional encoder having a coding rate of R= 1/3;
FIG. 12 is a simulation diagram illustrating a performance comparison among R=l/2 constraint codes for an R=V6 convolutional code; and
FIG. 13 is a simulation diagram illustrating a worst performance comparison among R=l/2 constraint codes using an R=1/6 convolutional encoder with a highest performance.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT
A preferred embodiment of the present invention will be described hereinbelow with reference to the accompanying drawings. In the following description, well known functions or constructions are not described in detail since they would obscure the invention in unnecessary detail.
A term "symbol" as used herein refers to a coded data bit output from an encoder. For convenience of explanation, it is assumed that the multicarrier communication system is a three-carrier CDMA communication system using three carriers.
- 10 -

In a communication system supporting both the IS-95 system and the multicarrier system, wherein transmission signals of the two different systems are overlaid at the same frequence bands, coded symbols are distributed such that a performance degradation may be minimized during decoding of the damaged symbols, and then the distributed coded bits are assigned to the respective carriers. Thus, even if one of the carriers has interference during reception, it is possible to perform decoding for only the coded bits transmitted via the other carriers, thereby improving the system performance.
Moreover, in the forward link, an R=l/6 convolutional code can be used for a channel encoder. Therefore, when the channel encoder generates R=1/6 convolutional codes, it is very difficult to find R=l/6 convolutional codes having a good decoding performance. Accordingly, the present invention is directed to generating R=l/6 convolutional codes with a good decoding performance and distributing the generated convolutional codes to multiple carriers. The R=/16 convolutional codes generated according to the present invention have a good performance in both a multicamer CDMA communication system and a DS-CDMA communication system,
A description will now be made regarding an operation of generating symbols for maximizing a channel performance and distributing the generated symbols in a CDMA communication system according to an embodiment of the present invention. For convenience, the present invention will be described hereinafter with reference to a multicarrier CDMA communication system
First, reference will be made to R-l/6 convolutional codes for a multicamer
- 11 -

CDMA communication system using three carriers. FIG. 6 illustrates a convolunonal encoder and a symbol distributer according to an embodiment of the present invention.
Referring to FIG. 6, a convolutional encoder 601 encodes one input data bit into six symbols which are allocated to three carriers A, B and C. For symbol allocation, a symbol distributer 602 uniformly distributes the six input bits to the three carriers by two bits. The symbol distributer 602 distributes the symbols output from the convolutional encoder 601, taking into consideration how many carriers are damaged out of the three carriers. By using this symbol distribution method, even though one or two out of the three carriers are damaged, performance degradation in channel decoding can be minimized.
A description will now be made as to a method of designing the symbol distributer 602. A bit error rate (BER) after channel decoding depends on a damaged portion for the symbols coded by a channel encoder. Therefore, even though the coded symbols are damaged, the symbols located at the minimized performance degraded portion are uniformly distributed to the carriers. Accordingly, even though the symbols for a certain channel arc all damaged, an increase in the BER after channel decoding can be minimized.
In addition, during transmission, the symbols output from a constituent encoder in the channel encoder are distributed to the carriers; during decoding, a constituent decoder in a channel decoder is selected such that the BER can be low even though the symbols for a certain carrier are all damaged.
- 12 -

Selection of the constituent decoder in the channel decoder is made in the following process. First, reference will be made to a convolutional code having a constraint length of K=9 and a rate of R=l/3. In the following descriptions, generator polynomials gi are represented by octal number. The convolutional code with K=9 and R= 1/3 has a free distance of dfm= 18. It is noted that there exists 5685 sett, when search is made for convolutional codes having K=9, R-l/3 and dfree 18, by changing generator polynomials g1, g2 and g3. Here, only non-catastrophic codes are selected. In addition, it is necessary to prevent the performance degradation even though a certain carrier is completely off, providing for application to the multicarrier system. From this point of view, it is preferable to maximize the free distance.
For a reference code for performance comparison, a convolutional code of (g1, g2 ,g3)= (557, 663, 711) is used which is used in the existing IS-95 system. In the IS-95 system, a free distance of the convolutional code is dfree=18, and free distances between constituent codes are dfree(g3S7,g661)=9, dfreer(g557,g711)=11,and dfree (g663,.g711)=10. A performance of a convolutional code can be predicted using a BER upper limit formula, which is determined by a transfer function.
For the IS-95 system, a transfer function of a convolutional code is T(D,1), = 5D18+7D20+O(D21), and a BER upper limit formula is 1 lD18+32D26+O(D21). When the convolutional code for the IS-95 system is view in the light of a constituent code, a catastrophic error propagation occurs at a combination of generator polynomials g, and g2. Therefore, when the convolutional codes for the IS-95 system are used for the multicarrier system, it is necessary to appropriately use interleaving and puncturing. Since the IS-95 convolutional codes
- 13 -

have the catastrophic error propagation in the light of the constituent codes, it is necessary to search for new convolutional codes suitable for the multicarrier system. For K=9, dfree(g1,g2)11. Here, not only the codes but also the constituent codes are non-catastrophic Since a first term of the BER upper limit formula is most influential, first and eighth codes can be considered to be most optimal codes. Here, since the pairs of first and eighth codes; second and seventh codes; third and fourth codes; and fifth and sixth codes are in reciprocal relation, they are the same codes essentially Therefore, there exist only four codes.
Table 1 is given to explain a characteristic of a convolutional encoder with K=9 and R=l/3.
[TABLE 1]

- 14 -

, 15 -

In Table 1, dI2 in a first term means d(447,543) and hereinafter, used in the same meaning. For information, when the codes are compared with the IS-95 codes in the light of the first term of the BER upper limit formula, the first and eighth codes are superior in performance to the IS-95 codes, the third, fourth, fifth and sixth codes are similar in performance to the IS-95 codes, and the second and seventh codes are inferior in performance to the IS-95 codes Therefore, it is preferable to use the eighth (or first) code.
In the meantime, there exist four or more codes whose the free distances among the constituent codes are 12, 12 and 10; among these codes, a generator polynomial for a superior code in the light of the first term of the BER upper limit formula is (g1,g2,g3) = (515, 567, 677). Shown in FIG. 10 is the simulation Tesult for performances of the convolutional code with R=l/3 in a AWGN (Additive White Gaussian Noise) environment in the case where the multicarrier (three-carrier) system has an optimal performance without damage of the respective carriers. In the following descriptions, the simulations of FIGs. 11-13 are all performed in the AWGN environment. represents a R=l/3 convolutional code for an existing IS-95 system, and and represent a R=l/3
- 16 -

convolutionsl code searched for in the above method.

A description will now be made regarding the case where the R=l/3 convolutional code is applied to the three-carrier system and a certain one of the three carriers is damaged (or lost). Although the original coding rate is 1/3, the loss of one carrier causes the coding rate to be equal to 1/2. Therefore, shown in FIG 11 is the simulation results for the 1/2 convolutional codes using the generator polynomials for the 1/3 convolutional codes. In FIG. 11, the respective conditions can be explained by the following through . FIG. 11 illustrates the worst performance graph for the R=l/2 convolutional codes using the generator polynomials for the R=l/3 convolutional code.
Optimal 1/2 convolutional code - g,=56l, g2=753, dfree(g1.g2)=12 the worst performance, g,=557, g2=711 out of three R=l/2
convolutional codes using the generator polynomial (557, 663, 711) for a 1/3
convolutional code used for the IS-95 system - catastrophic error propagation
occurs
the worst performance, g,=731, g,=615 (dfree,(gl,g2)=11) for a
R=l/2 convolutional code using the generator polynomial (731, 615, 537) for a
R=l/3 convolutional code
- 17 -

the worst perforrnance, g,=567, g;=677 (dfree(gl,g1)=10) for a R=l/2 convolutional code using the generator polynomial (515, 567, 677) for a R=l/3 convolutional code
When one carrier is damaged in a three-carrier system using an R=l/3 convolutional code, the coding rate becomes to be equal to R=l/2. In this case, a symbol distribution method for the symbol distributer is found by appropriately distributing the original R=l/3 convolutional codes to the three carriers using the following symbol deleting matrixes, so as to minimize the performance degradation even though the coding rate becomes R=l/2. In the simplest method, the following two symbol deleting matrixes are generated. In the following symbol deleting matrixes, "0" means the case where a carrier to which the corresponding symbol is provided is damaged, and "I" means the case where the carrier to which the corresponding symbol is provided is not damaged. That is, this means the case that the symbols corresponding to "0" are all allocated to a certain carrier, which is damaged during transmission. Therefore, one of the following various patterns of the symbol deleting matrix is selected, which minimizes the performance degradation even though one carrier is damaged, and the symbol distributer 602 provides the symbols to the respective carriers using the selected pattern. The followings are symbol deleting matrixes for finding a pattern used for the symbol distributer 602.

- 18 -

Further an m-sequence of length=8 is generated over a two-stage GF(3) using an m-sequence. For a ninth convolutional code, a sequence (1,2,0,2,2,1,0,1,2) is generated and then, the following symbol deleting matrix Ds is generated using the sequence.

Further, the following symbol deleting matrixes D4 and D5 are generated by
*
changing the row of the symbol deleting matrix D3

In addition, a sequence {2,1,0,1,1,0,1.2,1,0,0,0,2,1,2} is obtained by generating 15 random numbers over GF(3) using a random number, and the
- 19 -

following symbol deleting matrix D4 is created using the above sequence

Also, the following symbol deleting matrixes D7 and D8 are generated by changing the rows as in the method using the m-sequence

Next, a description will be made regarding a convolutional code having a symbol rate of 1/6. A K=9, R=l/6 convolutional code has a free distance of dfree=37. In searching for the convolutional codes having a free distance of dfree=37 by randomly changing the generator polynomials g1 g2 ..., g3. the following conditions should be satisfied
First, it should be a R=l/6 convolutional code with a good decoding performance.
Second, it should be a R=l/4 convolutional code with a good decoding
- 20 -

performance which has generator polynomials (g1,g2,g3,g4), (g1,g2,g5,g6) and (g1,g2,g5,g6), considering the case that one of three carriers is damaged in the three-carrier system.
Third, it should be a 1/2 convolutional code with a good decoding performance which has generator polynomials (g1,g2), (g3.g4 and (g5,g6), considering the case that two of three carriers are damaged in the three-carrier system.
In the second and third conditions out of the above three conditions, the performance degradation is minimized even though one or two of the three carriers are completely off, providing for the multicarrier system in which six output bits of the covolutional code are allocated to three carriers by two bits. From this point of view, it is preferable that the R=l/4 convolutional code and the R=1/2 convolutional code have the maximum free distance.
A method of searching for a R=l/2 convolutional code satisfying the third condition becomes apparent from the following description. There exist 35 non-catastrophic convolutional codes with R=l/2, K=9 and dfree=12. An upper limit formula for the BER is give as follows, and a coefficient c12 of the most important term D12 in determining the BER ranges from 33 to 123.

First, for the R=l/6 convolutional codes, there exist 180 R=l/6 convolutional codes with dfree=37, satisfying the third condition. It is assumed that
- 21 -

dfree(g1,g2,g5,g6)=12. Here, there exist 58 convolutional codes in which the first term of the BER upper limit formula for the R=l/6 convolutional code has a coefficient of C37=l. The following are the R=l/6 convolutional codes selected among the 58 convolutional codes after performance verification.

- 22 -
The following are 5 R=1/6 convolutional codes with a good decoding performance selected among the 12 performance-verified l/6 convolutional codes.

A performance of the R-1/2 convolunonal codes using five generator polynomials for the R=1/6 convolutional code are verified, and further, a performance of the R= 1/4 convolutional codes using five generator polynomials for the R-1/6 convolutional code are verified. First, a transfer function for the R=l/2 convolutional codes will be described with reference to Table 2 in which the generator polynomials are represented by octal number.
[TABLE 2]

No 1/2 Generator Polynomial BER Upper Limit Formula, Coeff c12, c12, c14 1 435,657 (33,0,281) 2 561,753 (33,0,281) 3 515,677 (38, 106,238) 4 545, 773 (38, 106,238) 5 463, 755 (38,0,274) 6 557,631 (38,0,274) 7 557,751 (40,33, 196) 8 .457,755 (40,33, 196) 9 453,755 (40,0,271) 10 :557, 651 (40,0,271) 11 471,673 (50,0,298) 12 537,615 (50,0,360)
- 23 -

- 24 -

A R=1/2 convolutional code with the highest performance is searched for by verifying the perfonnances of the respective R=1/2 convolutional codes in Table 2. In addition, performances of the R=l/2 convolutional codes are compared with performances of the optimal R=l/2 convolutional code used for the IS-95 system.

A performance comparison among the respective cases is shown in FIG. 12 FIG. 12 illustrates a performance comparison among R=l/2 constituent codes for the R=l/6 convolutional code. It is noted that the R=l/2 constituent codes for the R=l/6 convolutional code arc similar in performance to the optimal R=l/2
- 25 -

convolutional code.
Table 3 illustrates transfer functions for the R=1/6 convolimonal codes.
- 26 -
[TABLE 3]

- 27 -

- 2B ¦

The worst performances of the R=l/2 constituent codes using 5 R=1/6 cnnvolutional codes with the good decoding performance are as follows, with reference to Table 3.
the worst performance of a R=l/6 convolutional code (NO=1) having generator polynomials of (457, 755, 551, 637, 523, 727), - (523, 727)g, cl2 = 68
the worst performance of a R=l/6 convolutional code (NO-7) having generator polynomials of (515, 677,453, 755, 551, 717), - (515,677),. cl2 = 38
the worst performance of a R=l/6 convolutional code (NO=8) having generator polynomials of (545, 773, 557, 651, 455, 747)g - (545, 773),, cl2 = 38
the worst performance of a R=l/6 convolutional code (NO=11) having generator polynomials of (551, 677, 557, 651, 455, 747)R - (551, 677)g. c12 ,
- 29 -

= 38
the worst performance of a R=l/6 convolutional code (NO=29) having generator polynomials of (515, 677, 455,747,531,657), - (515, 677),, c12 = 38
The wont performances for the R= 1/4 constituent codes are as follows using the R=1/6 convolutional codes whose performances are verified for the R=l/2 constituent codes.
the worst performance of s R=1/6 convoltional code (NO-1) having generator polynomials Of (457, 755, 551,637,523, 727), - (551,637, 523, 727)g, c24 = 5
the worst performance of a R-l/6 convolutional code (NO=7) having generator polynomials of (515, 677,453, 755, 551, 717), - (515,677, 551
the worst performance of a R=)/6 convoiutional code (NO=8) having generator polynomials of (545, 773, 557,651, 455, 747), - (545, 773, 455,
the worst performance of a R=l/6 convolutional code the worst performance of a R=l/6 convolutional code (NO=29) having generator polynomials of (515, 677,455,747, 531,657), - (5 15, 677, 531,
FIG, 13 iilustrates a comparison among the worst performances of R= F/2
- 30 -

constituent codes using R=l/6 convolutional code with the highest performance.
The following are two R=l/6 convolutional codes with the good decoding performance, selected among the R=l/6 convolutional codes whose performances are verified for various cases in the above manners.

- 31 -
Further, to search for a symbol deleting pattern used for the three-carrier system, various symbol deleting matrixes are considered for the situation where one carrier is damaged, i.e., where the R=l/6 convolutional codes change to R=l/4 convolutional codes. The reason for searching for the symbol deleting matrix pattern is the same as described for the R=l/3 convolutional codes. The following matrixes can be used as a symbol deleting matrix pattern for a method of distributing symbols for R=l/6 convolutional codes.

-32-

- 33 -
Taking into consideration the case where two carriers are damaged in the three-carrier system, the following symbol deleting matrix pattern can be used in a method of distributing symbols for R=1/2 symbol-deleted convolutional codes using generator polynomials for the R=l/6 convolutional codes with a good decoding performance.

Turning to FIG. 6, there is shown the convolutional encoder 601 and the
- 34 -

symbol distributer 602 according to an embodiment of the present invention. In the exemplary embodiment, the convolutional encoder 601 has a coding rate of R=l/6 and uses generator polynomials of (545, 773, 557, 651, 455, 747). The detailed structure of the R=l/6 convolutional is illustrated in FIG. 7.
Referring to FTG. 7, upon receipt of input data, delays 711 -A to 711-H delay the input data bits sequentially. During the sequential delay of the input data bits, exclusive OR gates 721-A to 721-F output coded symbols. The coded symbols of FIG. 7 are provided to the symbol distributer 602 having the structure of FIG. 8
Referring to FIG. 8, the symbol distributer 602 is implemented by switches 811-A and 811-B. In FIG 8, when a symbol rate of a clock for controlling the switches 811-A and 811-B is over six times a symbol rate of the symbol distributer 602, the symbols can be distributed without symbol loss. That is, the switch 811-A
sequentially receives input symbols g1, g2, g3, g4, g5, g6, g1, g2, g3 and the switch
811-R distributes the input symbols to output nodes C1, c2, c3, c4 c5 and c6
FIG. 9 illustrates a transmission scheme including the channel encoder 601 and the symbol distributer 602 of FIG. 6
Referring to FIG. 9, first to fourth CRC generators 911-914 add CRC data in a specified number of bits to input data. Specifically, 12-bit CRC is added to the 172-bit data of the first rate; 8-bit CRC is added to the 80-bit data of the second rate; 6-bit CRC is added to the 40-bit data of the third rate; and 6-bit CRC is added to the 16-bit data of the fourth rate. First to fourth tail bit generators 921-924 add 8 tail bits to the CRC-added data. Therefore, the first tail bit generator 921 outputs
- 35 -

192 bits; the second tail bit generator 922 outputs 96 bits; the third tail bit generator 923 outputs 54 bits; and the fourth tail bit generator 924 outputs 30 bits.
First to fourth encoders 931 -934 encode data output from the first to fourth tail bit generators 921-924, respectively. Here, a K=9, R=l/6 convolutional encoder can be used for the encoders 931-934. In this case, the first encoder 931 encodes the 192-bit data output from the first tail bit generator 921 into 1152 symbols of full rate; the second encoder 932 encodes the 96-bit data output from the second tail bit generator 922 into 576 symbols of 1/2 rate; the third encoder 933 encodes the 54-bit data output from the third tail bit generator 923 into 324 symbols of about 1/4 rate; and the fourth encoder 934 encodes the 30-bit data output from the fourth tail bit generator 924 into 180 symbols of about 1/8 rate.
First to fourth symbol distributers 941-944 distribute the symbols output from the encoders 931-934, respectively. Here, for symbol distribution, a channel controller (not shown) generates control signals for distributing the channel coded bits such that the performance degradation may be minimized during decoding of received damaged bits, when the coded symbols are transmitted being overlaid on the symbols of a different system at the same frequency band. The symbol distributers 941-944 then assign the symbols output from the encoders 931-934 to the corresponding carriers according to the control signals, respectively.
Rate matchers 951 -953 each include a symbol repeater and a symbol deletion device. The rate matchers 951-953 match rates of the symbols output from the corresponding symbol distributers 942-944 to a rate of the symbols output from the symbol distributer 941. First to fourth channel interleavers 961-964 interleave the
- 36 -

symbols output from the symbol distributer 941 and the rate matchers 951-953, respectively.
For the DS-CDMA communication system, the symbol distributers 941-944 of FIG, 9 can be removed.
As described above, in the multicarrier system employing the frequency overlay method, the respective carriers have limited transmission powers according to the loading in the frequency bands of the existing 1S-95 system, which results in loss of data received at one or more carrier frequency bands. To solve this problem, by using the generator polynomials for the channel encoder and a symbol distribution method, it is possible to provide a high coding gain against the data loss due to the carrier loss, thereby preventing degradation of the BER.
While the invention has been shown and described with reference to a certain preferred embodiment thereof, it will be understood by those skilled in the art that various changes in forrn and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
- 37 -

-38-
We Claim
1. A channel transmission device for a multi-carrier CDMA communication
system using at least two carriers, comprising:
a channel encoder (601, 931-934) for encoding channel data to be transmitted into symbols at a predetermined coding rate; and a symbol distributor (602, 941-944) for receiving the symbols and distributing received symbols to the carriers according to a predetermined symbol deleting matrix pattern, wherein the symbol deleting matrix pattern is so determined as to distribute the symbols to the respective carriers with a minimized performance degradation even though a specific carrier is damaged.
2. The channel transmission device as claimed in claim 1, further
comprising:
a channel interieaver (961-964) for channel interleaving the distributed
symbols;
a demultiplexer (93) for distributing the interleaved symbols to the
carriers,
a plurality of orthogonal modulators (61-63) for generating orthogonally
modulated signals by multiplying the distributed symbols by orthogonal
codes for the corresponding channels;
a plurality of spreaders (71-73) for receiving the orthogonally modulated
signals and generating spread signals by multiplying the received
orthogonally modulated signals by a spreading code; and
a plurality of transmitters (81-83) for receiving the spread signals and
transmitting the received spread signals using the carriers.

-39-
3. The channel transmission device as claimed in claim 1 or 2, wherein the
channel encoder is a convolutional encoder with a coding rate of R=l/6.
4. The channel transmission device as claimed in one of claims 1 to 3,
wherein the symbol distributor comprises:
a first selector (811-A) for sequentially multiplexing the received symbols; and
a second selector (811-B) for distributing the multiplexed symbols to the carriers according to the symbol distribution signal.
5. The channel transmission device as claimed in one of claims 1 to 4,
further comprising:
a channel controller for generating a symbol distribution signal according to the predetermined symbol deleting matrix pattern, wherein the symbol distributor is arranged for distributing the received symbols to the carriers according to the symbol distribution signal.
6. A channel transmission method for a CDMA communication system using
at least two carriers, comprising the steps of:
encoding channel data to be transmitted into symbols at a predetermined coding rate; and
receiving the symbols and distributing received symbols to the carriers according to a predetermined symbol deleting matrix pattern, wherein the symbol deleting matrix pattern is so determined as to distribute the symbols to the respective carriers with a minimized performance degradation even though a specific carrier is damaged.

-40-
7. The channel transmission method as claimed in claim 6, wherein the
channel data is encoded using a convolutional encoder with a coding
rate of R=1/6.
8. The channel transmission method as claimed in claim 6 or 7, wherein the
symbol distribution step comprises the step of:
sequentially multiplexing the received symbols; and
distributing the multiplexed symbols to the carriers according to the
symbol distribution signal.
9. The channel transmission device as claimed in claim 1, wherein the
channel encoding device comprising;
a plurality of delays for delaying an input data bit to generate first to
eight delayed data bits;
a first operator for exclusively ORing the input data bit and the third,
fifth, sixth, seventh and eighth delayed data bits to generate a first
symbol;
a second operator for exclusively ORing the input data bit and the first,
second, third, fifth, sixth and eighth delayed data bits to generate a
second symbol;
a third operator for exclusively ORing the input data bit and the second,
third, fifth and eighth delayed data bits to generate a third symbol;
a fourth operator for exclusively ORing the input data bit and the first,
fourth, fifth, sixth, seventh and eighth delayed data bits to generate a
fourth symbol;

-41-
a fifth operator for exclusively ORing the input data bit and the first, fourth, sixth and eighth delayed data bits to generate a fifth symbol; and a sixth operator for exclusively ORing the input data bit and the first, second, fourth, sixth, seventh and eighth delayed data bits to generate a sixth symbol.
10. The channel transmission method as claimed in claim 6, wherein the channel encoding method for a CDMA communication system, comprising the steps of:
shiftingly delaying an input data bit to generate first to eight delayed data bits;
exclusively ORing the input data bit and the third, fifth, sixth, seventh and eighth delayed data bits to generate a first symbol; exclusively ORing the input data bit and the first, second, third, fifth, sixth and eighth delayed data bits to generate a second symbol; exclusively ORing the input data bit and the second, third, fifth and eighth delayed data bits to generate a third symbol; exclusively ORing the input data bit and the first, fourth, fifth, sixth, seventh and eighth delayed data bits to generate a fourth symbol; exclusively ORing the input data bit and the first, fourth, sixth, and eighth delayed data bits to generate a fifth symbol; and exclusively ORing the input data bit and the first, second, fourth, sixth, seventh and eighth delayed data bits to generate a sixth symbol.
A channel transmission device for a code division multiple access (CDMA) communication system using at least two carriers. In the channel transmission device, a channel encoder encodes channel data to be transmitted into symbols at a predetermined coding rate. A channel controller generates a symbol distribution signal according to a predetermined symbol deleting matrix pattern. The symbol deleting matrix pattern is so determined as to distribute the symbols to the respective carriers with a minimized performance degradation even though a specific carrier is damaged. A symbol distributer receives the symbols and distributes received symbols to the carriers according to the symbol distribution signal.

Documents:

00508-cal-1999-abstract.pdf

00508-cal-1999-claims.pdf

00508-cal-1999-correspondence.pdf

00508-cal-1999-description(complete).pdf

00508-cal-1999-drawings.pdf

00508-cal-1999-form-1.pdf

00508-cal-1999-form-18.pdf

00508-cal-1999-form-2.pdf

00508-cal-1999-form-3.pdf

00508-cal-1999-form-5.pdf

00508-cal-1999-g.p.a.pdf

00508-cal-1999-letters patent.pdf

00508-cal-1999-priority document others.pdf

00508-cal-1999-priority document.pdf

508-CAL-1999-FORM-27.pdf

« Previous Patent

Next Patent »

Patent Number

201698

Indian Patent Application Number

508/CAL/1999

PG Journal Number

06/2007

Publication Date

09-Feb-2007

Grant Date

09-Feb-2007

Date of Filing

31-May-1999

Name of Patentee

SAMSUNG ELECTRONICS CO.LTD.,

Applicant Address

416 MAETAN-DONG ,PALDAL-GU,SUWON-CITY KYUNGKI-DO

Inventors:

#	Inventor's Name	Inventor's Address
1	JONG-SEON NO	133,IMAE-DONG ,PUNTANG-GU,SONGNAM-SHI,KYONGGI-DO,
2	JAE-YOEL KIM	SANPON 9-DANJI BAEKDU APT.#960-1401,SANPON 2-DONG,KUNPO-SHI,KYONGGI-DO,
3	KYEONG-CHEOL YANG	50,YOUIDO-DONG,YONGUNGPO-GU,SEOUL

PCT International Classification Number

H03M 13/00

PCT International Application Number

N/A

PCT International Filing date

PCT Conventions:

#	PCT Application Number	Date of Convention	Priority Country
1	20141/1998	1998-05-30	Republic of Korea