AU3423799A - Method and an arrangement for modulating a signal - Google Patents

Description

WO 99/55048 PCT/FI99/00312 1 METHOD AND AN ARRANGEMENT FOR MODULATING A SIGNAL FIELD OF THE INVENTION The invention relates to a method for modulating a signal, the method using continuous phase modulation and comprising coding and fre 5 quency modulation of a signal to be transmitted. BACKGROUND OF THE INVENTION A modulation method used on a transmission path is a significant parameter when new data transmission systems are developed. Because of losses occurring on the transmission path and because of transmission path 10 capacity, data symbols to be transferred cannot be transmitted over the transmission path as such, but the symbols must be modulated using a suit able method so as to obtain good transmission path capacity and transmission quality. The bandwidth required by transmission is a significant factor par 15 ticularly in radio systems. The aim is to achieve maximum transmission capac ity while using a narrow bandwidth. On the other hand, the aim is to provide a transmitter and a receiver as easily and advantageously as possible. In radio systems, the aim is generally to use a modulation method having a constant envelope, because a C-class amplifier solution can then be used. The C-class 20 amplifiers are simple in structure and advantageous in efficiency. This is par ticularly relevant as far as terminal power consumption is concerned. There are several prior art modulation methods having a constant envelope, including Minimum Shift Keying MSK, Gaussian Minimum Shift Keying GMSK, Tamed Frequency Modulation TFM and Continuous Phase 25 Modulation CPM. The GMSK method is used in the GSM cellular radio sys tem. It has a narrow frequency spectrum and high performance, whereas data transmission rates are not very high. The coded CPM methods usually have a narrow frequency spectrum and high performance, making high data rates possible. However, equipment required become complex in structure, for 30 which reason these methods have not been used in prior art systems. BRIEF DESCRIPTION OF THE INVENTION An object of the invention is to provide a method and an arrange ment implying the method so as to enable high data rate transmission in a nar row frequency band without complex equipment required. This is achieved by WO 99/55048 PCTIFI99/00312 2 a method for modulating a signal, the method using M-level continuous phase modulation, wherein M can obtain values (2, 4, 8..) and symbols to be trans mitted comprise more than one bit, the method comprising coding and fre quency modulation of the signal to be transmitted. 5 The method of the invention is characterized by each signal to be transmitted being coded into a separate binary sequence, the sequences be ing transmitted binary-modulated so as to yield an M-level PSK constellation. The invention further relates to an arrangement for modulating a signal to be transmitted by M-level continuous phase modulation, wherein M 10 can obtain values (2, 4, 8..) and the symbols to be transmitted comprise more than one bit, the arrangement comprising a coder and a frequency modulator. The arrangement of the invention is characterized by comprising a coder (104) arranged to code each symbol to be transmitted into a separate binary sequence so as to yield an M-level PSK constellation. 15 The preferred embodiments of the invention are disclosed in the dependent claims. The basic idea of the invention is thus to provide a PSK constella tion, i.e. a state diagram, by means of binary modulation. The M-level symbols to be transmitted are coded into binary symbol sequences that are binary 20 modulated. The method and arrangement of the invention provide several ad vantages. The invention allows continuous phase modulation to be imple mented, which enables efficient frequency spectrum utilization and a relatively simple receiver structure compared with coded CPM methods, for example. 25 The capacity of the modulation method in accordance with the invention is good particularly when the signal-to-noise ratio is not very good. BRIEF DESCRIPTION OF THE DRAWINGS The invention is now described in closer detail in connection with the preferred embodiments with reference to the accompanying drawings, in 30 which Figure 1 illustrates a first example of an arrangement in accordance with the invention by means of a block diagram, Figure 2 illustrates a second example of the arrangement in accor dance with the invention by means of a block diagram, WO 99/55048 PCT/FI99/00312 3 Figure 3 illustrates a third example of the arrangement in accor dance with the invention by means of a block diagram, Figures 4a to 4c show examples of PSK state diagrams, Figure 5 illustrates a state diagram's state-to-state transitions of the 5 method in accordance with the invention, Figure 6 shows an example of the implementation of a coder in ac cordance with the invention, and Figure 7 shows an example of a coded bit stream. DETAILED DESCRIPTION OF THE INVENTION 10 Examine first an example of a preferred structure of an arrangement in accordance with the invention by means of a block diagram shown in Figure 1. The figure shows the structure of a terminal in a radio system in so far as relevant for the invention. Naturally, in order to function, a device to be imple mented should also comprise further components than those shown in Figure 15 1, as will be obvious to those skilled in the art. For the sake of clarity, however, these components are not dealt with in the figure and the description. For simplicity, let us here examine an example wherein symbols to be transmitted comprise two bits, in which case M = 2 (number of bits) = 4. The state diagram desired thus comprises four points. 20 The arrangement comprises a data source 100, which produces a digital signal 102 to be transmitted. The data source can be a microphone, for instance, connected to a speech coder, the signal to be transmitted thus being speech in a digital form. Other data sources may include a computer or a mo dem. Data bits are conveyed in parallel, i.e. two at a time, to a coder 104, 25 which, according to the invention, performs coding in which a two-bit symbol to be transmitted is presented by means of a binary symbol sequence. The cod ing will be described in closer detail below. In a preferred embodiment of the invention, the binary symbols thus obtained are conveyed to a filter 108 filtering the signal according to a spectral 30 pattern desired. A transfer function following the Gaussian distribution can preferably be selected as the transfer function of the filter. The transfer func tion can then be defined in the form g(t) = h(t)®rectIT WO 99/55048 PCTIFI99/00 3 12 4 where t is time, 0 represents convolution and a function rect(x) is defined by rect t) = when Iti< rect) =0 in other cases. 5 When the Gaussian distribution is used, a function h(t) can be se lected in the following manner: e ri(2) h(t)= where o-= ja BT =,6. ~2 oT 27TBT 10 Here, B is the 3-dB bandwidth of the filter with the impulse response h(t), T being thus the length of the data symbol. The signal thus obtained is further conveyed to a multiplier 110 to be multiplied by a factor h of the form %. The signal thus obtained is further conveyed to a frequency modulator 120 performing prior art frequency modu 15 lation by means of a voltage-controlled or a numerically controlled oscillator, for example. The phase of the modulated signal is of the form t-iT p(t')=Zah Jg(u)du 20 where a. = 1 - 2 * bs obtaining a value -1 or 1. The term bs com prises the binary sequence bits. The formation and form of the binary se quence are described below. A time reference t' is the beginning of the data to be transmitted. The modulated signal is further conveyed to radio frequency parts 25 122, which can be implemented according to the prior art. It is an advantage of the invention that the radio frequency parts of the GSM system, for example, can be used as the radio frequency parts. The modulated RF signal can be expressed in the form 30x(t') = cos(2nfOt'+p(t')+ p) WO 99/55048 PCT/FI99/00312 5 where EC is the energy of a modulating symbol, fo is the centre frequency, and <po is a random phase, which is constant for a period of one burst. In the radio frequency parts, a C-class amplifier can thus be used, which is a significant advantage particularly as far as portable terminals are concerned. 5 From the radio frequency parts the signal is conveyed to an an tenna 124. As the transfer function of the filter 108, a raised cosine-type func tion following a root raised cosine RRC function, for example, can also be ad vantageously selected. 10 Figure 2 illustrates a second embodiment of the invention. This em bodiment does not have a filter after the coder. In other respects, the solution is similar to the solution described above. Figure 3 illustrates a third alternative embodiment of the invention. In this alternative, the voltage-controlled oscillator of Figure 1 is replaced by an 15 integrator 300 and a phase modulator 302, from which the signal is further conveyed to the radio frequency parts. In other respects, the solution is similar to the one described above in connection with Figure 1. Examine next the coding of the invention performed in the coder 104 in closer detail. In the solution of the invention, coding is performed 20 wherein a symbol to be transmitted comprising several bits is presented by means of a binary symbol sequence. In this example, examine a case where M = 4. Hence, the symbol to be transmitted comprises two bits, and the sym bol can thus have four feasible values, for example values {0, 1, 2, 3}, the cor responding bit pairs being {00, 01, 10, 11}, for example. These four values are 25 thus to be transmitted by means of binary symbols. In the solution of the in vention, coding is performed wherein each two-bit symbol is presented by means of a sequence comprising three binary symbols. The coder 104 per forms this modification, which can also be performed by a computer, for ex ample. 30 Figure 4a shows an example of a feasible state diagram of the modulation method of the invention when M = 4. Since a modulation method with a constant amplitude is here concerned, the transitions of the state dia gram form a unit circle. The origins and terminals of the transitions are indi cated by points on the unit circle. The points are placed at regular intervals n/2 35 phase difference away from each other. There are four points, one for each feasible symbol {00, 01, 10, 11}. The receiver aims at interpreting transitions WO 99/55048 PCT/FI99/00312 6 from one state to another from a signal received. For example, a transition from state 00 to state 01 can take place by a transition 400, and to state 11 by means of a transition 402. In the prior art modulation methods the transitions are implemented in one step in a manner shown by Figure 4a. The modulation 5 method's susceptibility to errors is indicated by the Euclidean distance be tween adjacent points. In the solution of the invention, symbols to be transmitted compris ing more than one bit are presented by means of binary symbol sequences. When M = 4, the number of bits is two, and the length of a binary symbol se 10 quence representing one symbol is three bits. In the state diagram this means that each state-to-state transition, the transition 00 -- + 01 for example, is im plemented employing three transitions, as illustrated in Figure 4b, in which the transition from state 00 to state 01 is performed by means of three transitions 404, 406 and 408. 15 From each four points of the state diagram, a transition from one point to another can take place by means of three transitions, a 3-bit-long bi nary symbol sequence corresponding to these transition alternatives. Natu rally, transition combinations are numerous, as illustrated by Figure 4c by way of example. In Figure 4c, the transition from state 00 to state 01 takes place by 20 means of three different transitions 410, 412, 414. Despite the fact that the origins and terminals are the same, the path generated by the transitions dif fers from the one shown in Figure 4b. A different binary symbol sequence cor responds to the transitions of Figure 4c than to the transitions of Figure 4b. In order for the coding to be unambiguous, each state-to-state transition should 25 be chosen a particular binary symbol sequence. It should be noted here that when a signal modulated in accordance with the invention is received, the receiver multiplies the received signal be 4 4 tween the symbols either by a value e or e , where k = 0, 1, 2,... This makes the state diagram rotate by half the symbol interval (it/4). The rotation 30 takes place either counter-clockwise or clockwise, depending on the sign of the exponent. In the example described above, bit '1' is coded into value '-1', resulting in a counter clockwise transition. When said rotation by multiplier e 4 is taken into account, rotation in the state diagram always takes place clockwise. Binary combination '000' keeps the constellation stationary, and 35 correspondingly, '111' results in a % rotation on the circle. The exponential WO 99/55048 PCT/FI99/00312 7 multiplier thus causes the constellation either to be stationary or to move in one direction, never in both directions during one state transition. The state diagram transitions of the modulation method in accor dance with the invention can be presented as a three-dimensional path ac 5 cording to Figure 5. The figure illustrates two examples of a path comprising three transitions when the transition takes place from one state diagram point to another. A first path 508 It is indicated by a broken line, corresponding to the path shown in Figure 4b. A second path 510 denotes the transition from point 00 to point 11, being indicated by a line of dots and dashes. In an initial 10 situation 500 the paths diverge, and via steps 502 and 504, terminate at dif ferent points in step 506. Four different paths exist from each point (state), ending up at four different points. Susceptibility to transfer errors of the modulation method of the invention is thus not indicated by the Euclidean distance between the 15 points but by the Euclidean distance between the paths. In the method of the invention, coding can be implemented by means of a computer. A block diagram in Figure 6 illustrates a feasible em bodiment of a coder of the invention. Parallel-mode bits 102 of an M-level symbol to be coded are supplied as input to the coder. Each bit is conveyed to 20 the coder both directly and delayed in delay parts 600 to 604. In general, the number of the symbol bits is a two-based M logarithm, i.e. lo9 2 (M). Typically, M is 4, 8, 16.... A coder 606 comprises a state machine having M number of bi nary symbol sequences, the length of each being M-1, and one being selected for the output on account of the bits in the input. The output can be imple 25 mented in parallel mode, the output thus having M-1 lines 608, which are con verted into serial mode by reading them sequentially with a switch 610. Natu rally, converting into serial mode can also be implemented in other manners known. In the examples given above M = 4, in which case the number of lines 102 is two and the number of the output lines 608 is three. Correspondingly, if 30 M = 8, the number of the lines 102 is three and the number of the output lines 608 is seven. The coder of the invention can preferably be implemented by soft ware by using a signal processor or a general purpose processor. Examine an example of feasible binary symbol sequences for dif 35 ferent state-to-state transitions. The present example examines a case where M = 4. Consequently, a symbol to be transmitted thus comprises two bits, the WO 99/55048 PCT/F199/00312 8 symbol then being capable of having four feasible values, for example values {0, 1, 2, 3}, the corresponding bit pairs being {00, 01, 10, 11}, for example. The table has a binary symbol sequence corresponding to each transition. In total, there are four different sequences, and a transition from one point to another 5 can take place by any one of the four sequences. In the present example, the sequences are S, = {000, 010, 101, 111}. The rightmost column in the table indicates the index of the sequence. Transition Sequence S 0-+0 000 Si(1) 0111 S4) 0-+2 010 Si(2) 0 -3 101 Si(3) 1 0 010 Si(2) 1 -+ 1 000 Si(1) 1 -+2 101 Si(3) 1 -+3 111 S(4) 2 -111 Si(4) 2 - 1 101 Si(3) 2 2 000 Si(1) 2 -3 010 S(2) 3-+ 0 101 Si(3) 3 -1 010 Si(2) 3111 S(4) 3 -+ 3 000 Si(1) Table 1 10 Another alternative to a sequence is S 2 =000, 100, 011, 111}. The sequences should preferably be selected such that the number of '1' bits grows by one when a transition from a sequence to another takes place. It is to be noted that the rotation of the state diagram by the value e j where k = 0, 1, 2,.. at the receiver has been taken into account in the above sequences. 15 Hence, coding is performed in such a manner that the current state and the state to which a transition is being performed are determined, after which a binary sequence corresponding to the transition obtained is looked up in the table, the binary sequence then being set in the output of the coder.

WO 99/55048 PCT/FI99/00312 9 Figure 7 further illustrates a bit stream after the coding. The bit stream comprises a group of binary sequences 700 to 704, each binary se quence corresponding to one symbol to be transmitted. These binary se quences are transmitted sequentially in a time slot or in a frame, depending on 5 the multiple access method used. The solution of the invention can preferably be applied to any digital data transfer system, to a cellular radio system, subscriber terminal equipment and base stations, for example. Although the invention has been described above with reference to 10 the example of the accompanying drawings, it is obvious that the invention is not restricted thereto but it can be modified in many ways within the scope of the inventive idea disclosed in the attached claims.

Claims (11)

1. A method for modulating a signal, the method using M-level con tinuos phase modulation, wherein M can obtain values (2, 4, 8..) and symbols to be transmitted comprise more than one bit, the method comprising coding 5 and frequency modulation of the signal to be transmitted, c h a r a c t e r i z e d by each symbol to be transmitted being coded into a separate binary sequence, the sequences being transmitted binary-modulated so as to yield an M-level PSK constellation.

2. A method as claimed in claim 1, c h a r a c t e r i z e d by each 10 symbol to be transmitted being presented by means of a predetermined, serial mode binary sequence.

3. A method as claimed in claim 1, characterized by the length of each binary sequence representing the symbol to be transmitted be ing M-1 bits. 15

4. A method as claimed in claim 2, c h a r a c t e r i z e d by each transition from one point to another of the PSK constellation being imple mented by means of several different transitions determined by the binary se quence bits.

5. A method as claimed in claim 4, characterized by the 20 point-to-point transition being implemented by means of M-1 transitions.

6. A method as claimed in claim 1, c h a r a c t e r i z e d by the bi nary sequences being selected such that the number of '1' bits increases by one when a transition from one sequence to another takes place.

7. An arrangement for modulating a signal to be transmitted by M 25 level continuous phase modulation, wherein M can obtain values (2, 4, 8..) and the symbols to be transmitted comprise more than one bit, the arrangement comprising a coder (104) and a frequency modulator (120), c h a r a c t e r i z e d by the arrangement comprising the coder (104) arranged to code each symbol to be transmitted into a separate binary sequence so as to yield an M 30 level PSK constellation.

8. An arrangement as claimed in claim 7, c h a r a c t e r i z e d by the arrangement comprising the coder (104) arranged to code each symbol to be transmitted into a binary sequence whose length is M-1 bits. WO 99/55048 PCT/F199/00312 11

9. An arrangement as claimed in claim 7, c h a r a c t e r i z e d by the frequency modulator (120) being implemented by means of a voltage controlled oscillator.

10. An arrangement as claimed in claim 7, c h a r a c t e r i z e d by 5 the arrangement comprising a filter (108), which is operatively connected to the output of the coder.

11. An arrangement as claimed in claim 7, c h a r a c t e r i z e d by the coder (104) being implemented by software as a processor software.