DE19939588A1 - Device and method for transmitting a quadrature amplitude modulated transmission signal - Google Patents

Abstract

Device for sending a QAM-transmission signal comprising a trigger frequency f0, whereby the device is composed of one coding device (9) which codes data to form complex valued pairs (a, b) consisting of two data; a sensing device (14) for sensing complex valued pair signals with a first scanning frequency; a selector (36) which selects the carrier frequency f0; a memory (31) from which a phase increment delta phi associated to a carrier frequency can be read; a battery (25) which generates a phase signal phi depending on the phase increment delta phi reading, and at least one computation device (24) to produce the QAM-transmission signal by multiplying the complex value pairs with a phase rotation operator, whereby said complex valued pair is rotated at a phase angle phi corresponding to the phase signal.

Description

The invention relates to an apparatus and a method for Sending a quadrature amplitude modulated transmit signal (QAM transmit signal) with a specific carrier frequency, esp especially in a return channel for cable television Transmission systems.

Television signals are increasingly coming from a central television station via a broadband cable network to end devices of television transfer participants. Originally the television signa le only in one direction from the television station to the Transfer participants analog. So that the existing width ribbon cable network also for interactive services, especially in ternet services that can be used, it is necessary agile, next to the down channel, which is in a frequency spectrum ranges from about 90 MHz to 800 MHz, a return channel be to sit in which digital data from the participant the service providers can be sent. On the The forward channel and the return channel is a bidirectional communication tion between the service provider and the subscriber, connected to the broadband cable network. The Participants can, for example, digital over the return channel Commands to select movies or video games on the Service providers send or buy orders to chew send goods. The bidirectional communication via the up and down channel also enables an interacti ve learning about the broadband cable network, for example the Be answer transferred questions.

The frequency ranges for the are from the cable network operators The forward and return channels have been standardized. For the canal is a frequency range of approximately 90 to 800 MHz and for the Back channel a frequency range of about 5 to 65 MHz festge  has been laid. As a modulation type for the transmission of the data is the Quadrature Amplitude Modulation (QAM) method been.

With quadrature amplitude modulation, the modulation takes place the amplitude and frequency or phase of a harmonic vibration through two different time functions.

Fig. 1 shows a block diagram of a conventional QAM transmitter according to the prior art.

The data to be transmitted are fed to an encoder, which summarizes a sequence of binary data and ei assigned to a pair of numbers (a, b). The stock of values for the number lenpaare determines the exact type of modulation. Become for example, only four pairs of values ((1, 1), (1, -1), (-1, 1), (-1, -1)) is used, it is a pure phase modulation (QPSH). Take the two values a and b of the number lenpaares several amplitude values, so it acts is a QAM modulation scheme. The pair of numbers (a, b) can be understood as a complex number c = a + jb.

In the conventional QAM transmitter shown in FIG. 1, the complex data signal is sampled at a certain sampling rate and two identical low-pass filters (g (t)) are supplied to limit the band. The low-pass filtered data signals are then modulated by multiplication with carrier signals which are orthogonal to one another. The transmission signal is formed by summing the two signal components, ie the normal and quadrature components. The carrier frequency f _{0 of} the orthogonal carrier signals determines the spectral position of the transmission signal x (t). When used in a return channel for a broadband cable network, the spectral position of the transmission signal x (t) lies in a range from approximately 5 MHz to 65 MHz in accordance with the established standard.

In the QAM transmitter according to the prior art shown in FIG. 1, two analog mixing stages M1, M2 are necessary to generate the transmission signal. These mixing stages are technically complex in terms of circuitry, since the modulation or multiplication must be carried out at a very high frequency. The sampling frequency of the sampling switches S1, S2 shown in FIG. 1 must satisfy the sampling theorem. The frequency range of the return channel is between 5 MHz and approx. 65 MHz, so that the sampling frequency must be more than 100 MHz.

It is therefore the object of the present invention, a Device and method for sending a QAM To create transmission signal, in which the transmission signal with gerin according to the circuitry complexity directly in the carrier quenzlage is brought.

This object is achieved by a device for Sending a QAM transmit signal with in claim 1 given features and by a method for sending egg NES QAM transmission signal with those specified in claim 21 Features resolved.

The particular advantage of the invention is that none analog mixing stages for generating the QAM transmit signal not are agile.

The invention provides a device for transmitting a QAM transmit signal with a carrier frequency, the device comprising:
a coding device for coding data into complex-value data pairs each consisting of two data,
a sampling device for sampling the complex data pair signals with a first sampling frequency,
a selection device for selecting the carrier frequency,
a memory from which a phase increment assigned to the carrier frequency can be read out,
an accumulator for generating a phase signal as a function of a read phase increment,
at least one calculation device for generating the QAM transmit signal by multiplying the complex data pair by a phase rotation operator, the complex data pair being rotated by a phase angle corresponding to the phase signal.

In an advantageous development of the invention The sampled complex data pair becomes a device re filtered by digital low-pass filters to limit the band.

This offers the advantage that a large number of frequency bands that can be used simultaneously on a broadband cable.

In a further advantageous embodiment of the inventions device according to the invention are the filtered output signals of the digital low-pass filter by a further Abta stone direction sampled with a second sampling frequency.

In a further advantageous embodiment of the inventions The device according to the invention has a clocked th buffer for the temporary storage of the phase sensor gnals on.

In a further advantageous embodiment of the inventions According to the device according to the invention, the accumulator has an adder to add the read phase increment with the zwi stored phase signal of the previous clock cycle on.

In a further advantageous embodiment of the inventions device according to the invention, the accumulator has a modulo Calculation unit.  

In a further advantageous embodiment of the inventions The device according to the invention is the modulo calculation unit a two's complement calculation unit.

In a further advantageous embodiment of the inventions The device according to the invention is a plurality of calculation devices conditions for parallel data processing of a variety of com plex-valued data signals connected in parallel.

This offers the particular advantage that the data processing speed can be increased significantly.

In a further advantageous embodiment of the inventions device according to the invention, the carrier frequency of the QAM Transmission signal between 5 MHz and 65 MHz.

This offers the particular advantage that the invention Device as a transmitting device in a return channel Broadband television cable network can be used.

In a further advantageous embodiment of the inventions device according to the invention is the second sampling frequency a multiple of the first sampling frequency.

In a further advantageous embodiment of the inventions device according to the invention is the second sampling frequency more than 100 MHz.

In a further advantageous embodiment of the inventions The device according to the invention is the digital low-pass filter non-recursive filters.

In a further advantageous embodiment of the inventions Invention device are the digital low-pass filter In terpolation filter.  

In a further advantageous embodiment of the inventions The device according to the invention is the digital low-pass filter narrow band with an adjustable bandwidth that depends is adjustable from the symbol rate.

In a further advantageous embodiment of the inventions device according to the invention contains the calculation device at least one cordic calculation unit.

In a further advantageous embodiment of the device according to the invention, the cordless calculation unit has:
two shift registers for bit-wise shifting of the data of a complex data pair depending on an iteration number,
a first addition / subtraction device for addition / subtraction of the first date of the complex data pair with the shifted second date of the complex data pair depending on the sign of the phase angle to form the first date for the next iteration,
a second addition / subtraction device for addition / subtraction of the second date of the complex data pair with the shifted first date of the complex data pair depending on the sign of the phase angle to form the second date for the next iteration.

In a further advantageous embodiment of the inventions The device according to the invention contains the cordial calculation unit a third addition / subtraction device for Addition / subtraction of the phase signal with a Arcus tangent value depending on the sign the phase angle to form the phase signal for the next most iteration.  

In a further advantageous embodiment, the cordic calculation unit a decrement device for Decrement the iteration number.

In a further advantageous embodiment, the Be arithmetic unit a cordic calculation unit, which is fed back via a buffer.

In a further advantageous embodiment of the inventions device according to the invention has the calculation device several kordic calculations connected in series units.

The invention also provides a method for transmitting a QAM transmit signal with a specific carrier frequency, which has the following steps:

• a) Coding of binary data to complex data couple signals,
• b) sampling the complex data pair signals with egg its sampling frequency,
• c) Selecting a carrier frequency and reading an assigned ordered phase increments,
• d) generating a phase signal with a rotation phase in Ab dependence on the phase increment read,
• e) multiplying the complex data pair signals with a rotary operator that uses the data pair signals Generation of the QAM transmission signal around the rotational phases pha sends turns, whereby of the turned complex-valued Da use the real signal component det.

Preferred embodiments of the invention will become apparent with reference to the accompanying drawings for explanation tion described features essential to the invention.

Show it:  

FIG. 1 is a transmitter for generating a QAM transmission signal according to the prior art;

Fig. 2 is a block diagram showing a cable network connection in which the device according to the invention can be used;

Fig. 3 is a block diagram of the device according to the invention for transmitting a QAM transmit signal;

Fig. 4 shows a first embodiment of a cordial calculation unit according to the invention;

Fig. 5 shows a second embodiment of the cordless calculation unit according to the invention;

Fig. 6 is a flowchart showing the calculation process within a kordische calculation unit according to the invention.

Fig. 7 is a cordial calculation element according to the inven tion.

Fig. 2 is a block diagram showing a cable connection, in which the transmission device of the invention can be used.

Via a television broadband cable 1 , the cable television connection 2 of a subscriber is connected to a service provider via a data link channel in the frequency range from approximately 90 MHz to 800 MHz and via a data return channel in the frequency range from approximately 5 MHz to 65 MHz. The Ka bel television connection 2 is connected via a signal line 3 to a subscriber television cable modem 4 . The subscriber television cable modem 4 outputs an analog television signal to a television terminal 6 via an analog signal line 5 . Via a bidirectional digital data line 7 , the subscriber television cable modem 4 is connected to a digital terminal, in particular a special computer or PC 8 . The binary data emitted by a digital terminal 8 are sent by the transmission device according to the invention as a QAM transmission signal for the lines 3 , 1 to the service provider in a data return channel.

Fig. 3 shows a block diagram of the device according to the invention for transmitting a QAM transmit signal. Binary data are fed via the digital signal line 7 to a coding device 9 for coding data into complex-value data pair signals consisting of two data a, b. The signal output 10 for the first data signal is connected via a line 11 and the second output 12 for the second data signal is connected via a line 13 to a sampling device 14 for sampling the complex data pair signals with a first sampling frequency f _t . The sampled first data signal is fed via a line 15 and the second sampled data signal is fed via a line 16 to the digital low-pass filters 17 , 18 . The digital low-pass filters 17 , 18 are preferably non-recursive filters with a high sampling frequency of more than 100 MHz or alternatively cascade-shaped filters, the first cascade stage of which is a non-recursive filter and the second cascade stage of which is an interpolation filter, in particular a COMB -Filter with a sampling frequency of more than 100 MHz. The sampling frequency of the first cascade stage, ie the non-recursive filter, is lower than the sampling frequency of the second cascade stage, ie the interpolation filter. The digital low-pass filters 17 , 18 are narrow-band low-pass filters with a bandwidth that can be set as a function of the symbol rate.

The sampled and filtered first data signal arrives via a line 19 and the second sampled and filtered data signal arrives via a line 20 to a further sampling device 21 , which samples the signals with a sampling frequency f _a and via lines 22 , 23 of a cordless calculation device 24 to generate the QAM transmit signal on line 3 . The calculation device 24 performs a multiplication of the complex data pair present on the lines 22 , 23 by a rotary operator, where the complex data pair signal is rotated by a phase angle ϕ corresponding to a phase signal.

The phase signal is generated in an accumulator 25 , which supplies the phase signal via a line 26 to the calculation device 24 . The accumulator 25 contains a clocked buffer 27 for buffering the phase signal. The buffered phase signal is fed back via lines 28 , 29 to an adder 30 . The adder 30 adds the feedback buffered phase signal with a phase increment read out from a memory 31 . The memory 31 is connected via a line 32 to the second input of the adder 30 . The Summensi signal of the adder 30 is fed via a line 33 to a modulo calculation unit 34 which is connected on the output side to the buffer store 27 via a line 35 . The module calculation unit 34 is preferably a two's complement calculation unit. A desired carrier frequency f _{0 is} selected by means of a selection unit 36 and this is communicated to the memory 31 via a line 37 . The desired carrier frequency is characterized by a constant phase increment δϕ. In the memory 31 , an associated phase increment is assigned to each carrier frequency in a table. The phase increment associated with the carrier frequency f ₀ is read out via the readout line 32 and fed to the digital accumulator 25 . The adder 30 of the accumulator 25 adds the phase increment read out to the phase value ϕ of the previous clock cycle. If a certain phase threshold, for example π, is exceeded, the modulo calculation unit subtracts twice the value of the phase threshold, for example 2π. The phase range from -π to + π can be scaled or mapped to the range from -1 to +1 by appropriate scaling. When the digital accumulator 25 is implemented in a two's complement number representation, this modulo operation is carried out without additional circuitry. The iteratively formed phase signal ϕ _{i + 1} = ϕ _{i + δϕ} is supplied via the line 26 to the calculation device 24 , which performs a cordial calculation process. The cordless calculation device 24 has two data outputs 3 , 38 . The second data output 38 , at which the calculated imaginary part of the data output signal is present, is not connected.

The selectable by the selection device 36 carrier frequency is preferably between 5 MHz and 50 MHz.

The second sampling frequency f _{a of} the second sampling device 21 is preferably a multiple of the first sampling frequency f _t of the first sampling device 14 . In a preferred embodiment, the second sampling frequency f _a is more than 100 MHz. The first sampling frequency f _t corresponds, for example, to a symbol rate of 0.16-2.65 Mbaud.

Fig. 4 shows a first embodiment of the cordless Be calculation unit 24 , as shown in Fig. 3.

The cordless calculation unit 24 contains a cordial calculation component 40 with four inputs 41 , 42 , 43 , 44 and four outputs 45 , 46 , 47 , 48 . The cords calculation component 40 is connected to a fixed value memory, in particular a ROM memory 51 , via lines 49 , 50 . At the first input 41 , the first data signal of the complex data signal pair is applied via line 22 and the second data signal of the complex data signal pair is applied via line 23 to fourth input 44 . The accumulator 25 shown in FIG. 3 is connected to the second input 42 of the cordial calculation element 40 via the line 26 . The third connection 43 of the kordischen calculation element 40 receives an iteration number N via a line 52 , which indicates the number of iteration steps carried out.

At the output 45 , the first data of the complex data signal pair after an iteration step and at the fourth output 48 , the second data of the complex data signal pair after the iteration over lines 53 , 54 are applied to clocked registers 55 , 56 . At the output 46 , the phase signal for the next iteration is applied via a line 57 to a clocked buffer store 58 . The iteration number applied to the input 43 is decremented within the calculation element 40 and applied to a buffer 60 at the output 47 via a line 59 . After carrying out an adjustable number of iteration steps via line 3, the intermediate memory 55 emits the QAM transmission signal. During the execution of the iteration steps, the first datum of the complex data signal stored in the buffer 55 is fed back to the first input 41 via a feedback line 61 .

The intermediate memory 58 , in which the phase signal is temporarily stored, is connected via a feedback line 62 to the second input 42 of the kordische calculating element 40 .

The buffer memory 60 , in which the iteration number is temporarily stored, is fed back via a feedback line 63 to the third input 43 of the cordial calculation element 40 .

The intermediate store 56 is fed back via a feedback line 64 to the fourth input 44 of the cordial calculation element.

The feedback lines 61 , 62 , 63 , 64 are connected to the inputs 41 , 42 , 43 , 44 of the cordial calculation element 40 via controllable switches 65 , 66 , 67 , 68 . The dashed lines show in Fig. 4 the switching position of the scarf ter after performing the first iteration step.

Fig. 5 shows a second embodiment of the cordless calculation unit 24th In the embodiment shown in FIG. 5, several identical cordic calculation components 40-1 , 40-2 ,. , , 40- (n + 1) connected in series and form a so-called pipeline structure. The various KENEN calculation elements 40-1 , 40-2 ,. , , 40 (M + 1) are connected via a multiplexer / demultiplexer 69 via read and write lines 70 , 71 to a common read-only memory 72 for storing arc tangent values.

FIG. 6 shows a schematic flowchart of the calculation process within a cordial calculation element 40 , the internal structure of which is shown in FIG. 7.

In a first step S1, the two data a, b of the complex data input signal, the phase signal p and the number N of the iteration steps to be carried out are applied. The first date a is applied to the input 41 , the second date b to the input 44 , the phase signal to the input 42 and the iteration number N to the input 43 of the kordische calculation element 40 .

In a calculation step S2, the data a, b of the complex data signal pair are calculated iteratively depending on the sign of the phase signal.

a '= a - signum (ξ). (2 ⁿ . B)

b '= b + signum (ξ). (2 ^n.a )

In the calculation step S2, the phase signal for the next iteration step is also calculated:

ξ '= ξ - signum (ξ). arcus tangens ( 2 ⁿ )

The arc tangent value is read from the read-only memory 51 in the first embodiment shown in FIG. 4 and from the read-only memory 72 in the embodiment shown in FIG. 5.

The calculated values are fed back in a step S3 in the first embodiment shown in FIG. 4 to the inputs of the corded calculation element 40 or in the second embodiment shown in FIG. 5 to an identical corded calculation element connected downstream.

Steps S2, S3 are carried out until:

n = NM

In step S4, the first datum of the complex data signal is read out at output S3 and output as a QAM transmit signal via line 3 .

The following applies:

x '= k. Re {(x + jy) e ^{j ϕ} }

Fig. 7 shows the circuitry structure of a kordi's calculation element 40 in detail.

The calculation element 40 contains shift registers 73 , 74 for bit-wise shifting depending on the iteration number n, a bit-wise shift to the left if n <0 and a bit-wise shift to the right if n <0. The iteration number n is applied to the third input 43 of the calculation element 40 .

The calculation element 40 also contains a first addition / subtraction device 75 which, depending on the signs of the applied phase signal, adds or subtracts the second date b shifted bit by bit in the second shift register 74 with the first date a.

The cordic calculation element 40 further contains a second addition / subtraction device 76 which, depending on the sign of the applied phase signal, sums or subtracts the first date a bit-shifted bit by bit in the first shift register 73 with the second date b.

The kordische calculation element 40 also has a third addition / subtraction device 77 , which detects the phase signal present at the input 42 at the previous iteration depending on the sign of the phase signal with an arctangent value, which comes from a read-only memory via the line 51 is read out, summed or subtracted.

The cordial calculation element 40 finally comprises a decrementing device 78 which decrements the iteration number applied to the input 43 for the next iteration.

Reference list

1

broadband television cable

2nd

Cable TV connection

3rd

management

4th

Subscriber cable modem

5

TV signal line

6

TV

7

binary data line

8th

digital terminal

9

Coding device

10th

Output connector

11

Output line

12th

Output connector

13

Output line

14

Scanner

15

management

16

management

17th

digital filter

18th

digital filter

19th

management

20th

management

21

Scanner

22

management

23

management

24th

cordic calculation unit

25th

accumulator

26

management

27

Cache

28

management

29

Feedback line

30th

Adder

31

Read only memory

32

management

33

management

34

Modulo calculation unit

35

management

36

Selection device

37

management

38

not used

39

not used

40

cordic calculation element

41

entrance

42

entrance

43

entrance

44

entrance

45

output

46

output

47

output

48

output

49

management

50

management

51

Read only memory

52

management

53

management

54

management

55

Cache

56

Cache

57

management

58

Cache

59

management

60

Cache

61

Feedback line

62

Feedback line

63

Feedback line

64

Feedback line

65

switch

66

switch

67

switch

68

switch

69

Multiplexer / demultiplexer

70

management

71

management

72

Read only memory

73

Shift register

74

Shift register

75

Addition / subtraction facility

76

Addition / subtraction facility

77

Addition / subtraction facility

78

Decrement device

Claims (21)

1. Device for transmitting a QAM transmission signal with a carrier frequency f ₀ , the device comprising:
a coding device ( 9 ) for coding data for complex data pairs (a, b) consisting of two data;
sampling means ( 14 ) for sampling the complex data pair signals at a first sampling frequency;
a selection device ( 36 ) for selecting the carrier frequency f ₀ ;
an accumulator ( 25 ) for generating a phase signal ϕ as a function of a phase increment δϕ assigned to the carrier frequency; and
at least one calculation device ( 24 ) for generating the QAM transmission signal by multiplying the complex data pairs by a phase rotation operator, the complex data pair being rotated by a phase angle ϕ corresponding to the phase signal and the real part being released from the rotated complex data for further processing . 2. Device according to claim 1, characterized in that the sampled complex data pair signals are filtered by digital low-pass filters ( 17 , 18 ) for band limitation. 3. Apparatus according to claim 1 or 2, characterized in that the filtered output signals of the digital low-pass filter ( 17 , 18 ) are sampled by a further sampling device ( 21 ) with a second sampling frequency f _a . 4. Device according to one of the preceding claims, characterized in that the accumulator ( 25 ) has a clocked buffer ( 27 ) for buffering the phase signal. 5. Device according to one of the preceding claims, characterized in that the accumulator ( 25 ) has an adder ( 30 ) for adding the read phase increment with the buffered phase signal of the previous clock cycle. 6. Device according to one of the preceding claims, characterized in that the accumulator ( 25 ) has a modulo calculation unit ( 34 ). 7. Device according to one of the preceding claims, characterized in that the modulo calculation unit a two's complement Calculation unit is. 8. Device according to one of the preceding claims, characterized in that the computing means ( 24 ) for parallel data processing of a plurality of complex data signal pairs are connected in parallel. 9. Device according to one of the preceding claims, characterized in that the carrier frequency f _{0 is} between 5 MHz and 65 MHz. 10. Device according to one of the preceding claims, characterized in that the second sampling frequency f _{a is} a multiple of the first sampling frequency f _t . 11. Device according to one of the preceding claims, characterized in that the second sampling frequency f _{a is} more than 100 MHz. 12. Device according to one of the preceding claims, characterized in that the digital low-pass filter ( 17 , 18 ) are non-recursive filters. 13. Device according to one of the preceding claims 1 to 11, characterized in that the digital low-pass filter ( 17 , 18 ) are cascaded ausgebil det, wherein a first cascade stage is a non-recursive filter and a second cascade stage is an interpolation filter, the sampling frequency first cascade stage is lower than the sampling frequency of the second cascade stage. 14. Device according to one of the preceding claims, characterized in that the digital low-pass filter ( 17 , 18 ) are narrow-band with an adjustable bandwidth, the bandwidth depends on the symbol rate. 15. Device according to one of the preceding claims, characterized in that the calculation unit ( 24 ) has at least one cordic calculation element ( 40 ). 16. Device according to one of the preceding claims, characterized in that the calculation device ( 24 ) has a cordic calculation element ( 40 ) which is fed back via intermediate memories ( 55 , 56 , 58 , 60 ). 17. Device according to one of the preceding claims 1 to 15, characterized in that the calculation device ( 24 ) contains a plurality of series-connected cords calculation elements ( 40 ). 18. Device according to one of the preceding claims, characterized in that the cordial calculation element ( 40 ) has two shift registers ( 73 , 74 ) for shifting the two data of the complex data pair in dependence on an iteration number n. 19. Device according to one of the preceding claims, characterized in that the kordische calculation element ( 40 ) a first addition / subtraction device ( 75 ) for adding / subtracting the first date of the complex data pair with the bitwise se shifted second date of the complex data pair depending from the sign of the phase angle to form the first date for the next iteration and a second addition / subtraction device ( 76 ) to add / subtract the second date of the complex data pair with the bitwise shifted first date of the complex data pair depending on the sign of the phase angle to form the second datum for the next iteration. 20. Device according to one of the preceding claims, characterized in that the calculation element ( 40 ) has a decrementing device ( 78 ) for decrementing the applied iteration number for the next iteration. 21. A method for transmitting a QAM transmission signal with a specific carrier frequency, comprising the following steps:

• a) coding of binary data to complex data pair signals,
• b) sampling the complex data pair signals with a sampling frequency,
• c) selecting a carrier frequency and reading out an assigned phase increment,
• d) generating a phase signal with a rotational phase as a function of the phase increment read,
• e) Multiplying the complex data pair signals with a rotary operator, which turns the data pair signals to generate the QAM transmit signal around the rotation phases pha, the real signal portion being used by the rotated complex data pair signals.