WO2001035566A2

WO2001035566A2 - Data adaptation method and device

Info

Publication number: WO2001035566A2
Application number: PCT/DE2000/003341
Authority: WO
Inventors: Josef Laumen
Original assignee: Robert Bosch Gmbh
Priority date: 1999-11-10
Filing date: 2000-09-26
Publication date: 2001-05-17
Also published as: WO2001035566A3; DE19953894B4; DE19953894A1

Abstract

The invention relates to a data transmission method in which a digital information signal is transmitted over a physical channel (K) between a transmitter (S) and a receiver (E). The inventive method comprises the following steps: preparing, in transmitter (S), the digital information signal in the form of at least one first bit data block with a first bit data rate; forming, in transmitter (S), a second bit data block (B1, B2) from the bits of the first bit data block and from one or more repetitive bits with a second bit data rate; transmitting, from transmitter (S) to receiver (E), the second bit data block (B1, B2) over the physical channel (K) with the second bit data rate; forming, in receiver (E), a third bit data block from the bits of the transmitted second bit data block (B1', B2') with the second bit data rate, and; forming a fourth bit data block from the bits of the third bit data block with the first bit data rate. In the case of a group of received bit values, which is comprised of one bit and of one or more corresponding repetitive bits, it is decided which bit rate is adopted in the fourth bit data block according to a predetermined algorithm while taking at least two received bits into account.

Description

Data transmission method and device

STATE OF THE ART

The present invention relates to a data transmission method in which digital information signal is transmitted over a physical channel between a transmitter and a receiver, and a corresponding data transmission device.

Although applicable in principle to any data transmission, the present invention and the problem on which it is based are explained in relation to a cellular UMTS data transmission system (UMTS = Universal Mobile Telephone System).

In such data transmission systems, it is common for technical data not to be able to implement any data rate on the physical channel. Instead, only a few discrete values are permitted for the channel data rate to be transmitted, such as in the UMTS (Universal Mobile Telephone System) system. Channel data rates of 32 * 2 kbit / s, k = 0 ... 5, are specified for single code transmission, ie the discrete values correspond to 32 kbps, 64 kbps, 128 kbps, 256 kbps, 512 kbps and 1024 kbps. Nevertheless, information transfer with any source data rate should be supported by UMTS systems.

Furthermore, information to be transmitted, in particular in mobile communication systems, should be protected against transmission via the physical channel by means of a channel coding method in order to ensure reception that is as error-free as possible. In a UMTS system in particular, such channel coding is carried out either using convolutional codes or turbo codes and is usually decoded at the receiver using a Viterbi or a Maximum A Posteriori (MAP) decoder.

Channel coding methods of the current UMTS specification provide code rates R _co de von, 1/3 and when using convolution and turbo codes. With regard to the discrete channel data rates mentioned above, this means that only discrete source data rates of approx. R _CO de * 32 * 2 kbit / s would be permitted if no additional adaptation of the data rate were provided. If no additional rate adaptation had been provided, for example with code rate R _CO de ^ only source data rates of approx. 16, 32, 64, 128, 256 or 512 kbps would be permitted. However, in order to support any source data rates, the data rate (rate matching) of the channel-coded information is adapted to the discrete channel data rate after the channel coding. FIG. 3 is a schematic illustration of a data transmission device in which the method according to the invention can be used.

In Fig. 3, S denotes a transmitter, K a physical channel (e.g. a radio channel), E a receiver, B1 and B2 bit data blocks to be transmitted, Bl 'and B2' bit data blocks transmitted. 50 is a channel transmission device. The units necessary for their formation are shown in the respective bit data blocks B1 and B1. On the one hand, in the transmitter S, reference numeral 10 designates an information signal input device for inputting a digital information signal, 20 a channel encoder, 30 an interleaving or scrambling unit and 40 a rate adaptation device. On the other hand, in the receiver E, reference numeral 10 'denotes an information signal output device for outputting the transmitted digital information signal, 20' a channel decoder, 30 'a deinterleaving or design unit and 40' a rate adaptation device.

The common algorithm for adapting or adapting the data rate with devices 40 and 40 'provides for puncturing or repetition of individual bits (unequal repetition) in the data stream.

FIG. 4 is a schematic illustration of the common puncture and repetition rate adjustment techniques. In Fig. 4, P denotes a puncturing which forms NM bits from N bits and R a repetition which forms N + M bits from N bits. T ₀ is a normalization time interval, for example 1 s.

If individual bits of a data frame of constant duration are punctured out, i.e. deleted and not transmitted, the data rate on the channel drops, while when repeating bits, additional replicas of these individual, selected bits are embedded in the data stream, thus increasing the effective data rate on the channel. Any source data rate can thus be adapted to the specified channel data rates.

3 further shows that in the transmitter S the rate adjustment with the rate adjustment device 40 follows the channel coding with the channel coding device 20. Between the two devices there is only the interleaving unit 30 for scrambling the position of the bits in the data block B1, B2, etc.

In the receiver E, the decoder 30 ', in particular in the form of a convolution or turbo decoder, takes place when the rate adaptation is undone by means of the rate adaptation device 40'. Between the two devices, there is only one deinterleaving unit 30 'according to the structure in the transmitter S for descrambling the position of the bits in the data block B1', B2 ', etc. In the case of puncturing P, there is no information about the punctured bits in the receiver E. This "zero information" must be communicated to the following convolution / turbo decoder 20 ^' . How this "zero information" is actually realized is clearly predetermined by the convolution / turbo decoder 20 '.

In contrast to this, in the case of the unequal repetition R, two information values are available before the rate dematching of individual bits, and the question thus arises as to how a single meaningful value can be found from these two values.

The simplest and most obvious variant is to reject any redundancy and exactly such bits in the

To use receivers that occurred at the positions of the original bits before rate matching in the transmitter S. In principle, existing information is given away with this procedure.

ADVANTAGES OF THE INVENTION

The idea on which the present invention is based is that redundancy bits with repetition bit values used in the transmission are not simply discarded, but are evaluated together with the corresponding transmission bits. The invention shows how a single meaningful value is found in rate dematching from two or more different received bit values of the same original bit, so that the overall error protection of the transmission system is significantly improved compared to the trivial solution.

The data transmission method according to the invention with the features of claim 1 and the data transmission device according to claim 10 have the particular advantage that the bandwidth efficiency and / or capacity of the channel of this data transmission system (at least in mobile radio systems) is also improved by improving the error protection properties of the system , In mobile radio systems in particular, with improved error protection, the transmission power of a signal on the air interface can be reduced without influencing the reception quality. This enables an operator of cellular networks to increase the number of users of his network, i.e. to increase the channel capacity of the mobile network.

Advantageous further developments and improvements of the method according to the invention specified in claim 1 are found in the subclaims.

According to a preferred development, certain pairs of bits and repetition bits are transmitted, the maximum value of the amount in the case of a pair consisting of a bit and a corresponding repetition bit of both received bit values is determined and this is transferred to the fourth bit data block, while the "smaller" received bit value is rejected.

According to a further preferred development, certain pairs of bits and repetition bits are transmitted, the arithmetic mean of both reception bit values being determined in the case of a pair of reception bit values consisting of a bit and a corresponding repetition bit, and this is adopted in the fourth bit data block.

According to a further preferred development, certain pairs of bits and repetition bits are transmitted, wherein in the case of a pair consisting of bit “a” and a corresponding repetition bit “b”, the quadratic mean value ^ yes ² + b ² ) / 2 of both receive bit values is first determined to adjust the amount of the value to be transferred to the fourth bit data block and that the sign of the value to be transferred to the fourth bit data block is chosen as the sign of the largest received bit value.

According to a further preferred development, certain pairs of bits and repetition bits are transmitted, with in the case of a pair consisting of a bit “a” and a corresponding repetition bit “b” initially an average of any order W (\ a \ + \ b) / N of both Emp¬

catch bit values is determined, where | α | and

denote the amount of the respective received bit value and N = 3, 4, 5, ... to determine the amount of the value to be transferred to the fourth bit data block and that the sign to be transferred to the fourth bit data block as the sign of the amount largest receive bit value is selected.

According to a further preferred development, certain pairs of bits and repetition bits are transmitted, wherein in the case of a pair consisting of a bit and a corresponding repetition bit, an average value of any order W (| αj ± \ b \) IN is first determined, whereby

and denote the amount of the respective received bit value, a is the larger received bit value and N = 3, 4, 5, ..., the amount of the value to be transferred to the fourth bit data block increasing to - (| α | +

/ N is determined if the signs of a and b are identical, and the amount is determined to be ^ (| α | - \ b \) IN if the signs of both

Receiving bit values are different, and that the sign of the value to be transferred to the fourth bit data block is selected in accordance with the sign of the largest received bit value a. According to a further preferred development, the bits of the first bit data block are encoded before the formation of the second bit data block and the bits of the fourth bit data block are decoded accordingly.

According to a further preferred development, the bits of the first bit data block are scrambled before the formation of the second bit data block and the bits of the third bit data block are descrambled accordingly.

According to a further preferred development, the transmission is carried out in analog form, preferably via a radio link.

DRAWINGS

Embodiments of the invention are shown in the drawing and will be explained in more detail in the following description.

Show it:

FIG. 1 shows a schematic illustration of an embodiment of the data transmission method according to the invention;

FIG. 2 shows a graphic representation of the frame error rate or bit error rate in the case of the according to the procedure and the trivial procedure;

Figure 3 is a schematic representation of a data transmission device in which the inventive method is applicable; and

Figure 4 is a schematic representation of the usual

Rate adjustment techniques puncturing and re-petition.

DESCRIPTION OF THE EMBODIMENTS

In the figures, identical reference symbols describe identical or functionally identical elements.

FIG. 1 is a schematic illustration of an embodiment of the data transmission method according to the invention.

In Fig. 1, SlOO to S700 denote process steps. The present embodiment of the data transmission method according to the invention can run on the data transmission system illustrated in FIG. 3. Therefore, reference is made to the reference numerals of FIG. 3 in terms of apparatus.

In step S 100, the digital information signal is provided in the form of a first bit data via the information signal input device 10 in the transmitter S. blocks with a first bit data rate, for example 10 information bits, including possible taiibits for the subsequent channel encoder 20.

In step S200, coding is carried out at rate 1/3 of the first bit data block by the encoder 20 in the transmitter S. The 10 information bits become 30 coded bits. Furthermore, in step S200, scrambling takes place by means of the interleaving unit 30.

In step S300, a second bit data block B1 is formed from the bits of the coded first bit data block and repeat bits, with every third bit being repeated, at a second bit data rate via the rate adaptation device 40 in the transmitter S. As a result, there is now a second bit data block B2 with 40 bits.

In step S400, the second bit data block B1 is transmitted via the physical channel K at the second bit data rate from the transmitter S to the receiver E.

In step S500, a third bit data block is formed from the bits of the transmitted second bit data block B1 ', B2' at the second bit data rate in the receiver E by means of the rate adaptation device 40 '.

In step S600, a fourth bit data block is formed in the receiver E by means of the rate adaptation device 40 ′ from the bits of the third bit data block with the Most bit data rate, in the case of each pair consisting of a bit and a corresponding repetition bit the maximum value of the amount of both receive bit values is determined and this is transferred to the fourth bit data block, while the amount of the "smaller" receive bit value is discarded. In other words, every third and fourth bit of the 40 bits received is combined in this way in order to return to a number of 30 coded bits.

In step S700 a descrambling takes place by means of the deinterleaving unit 30 'and then a decoding of the first bit data block by the decoder 20' in the receiver E. The 30 coded bits in turn become the 10 information bits which were previously sent in the transmitter S.

According to a further exemplary embodiment, information with a source data rate of 31.6 kbit / s is to be transmitted via a UMTS system, error protection being ensured by means of a convolution code of the rate 1/3.

According to the UMTS specifications, the source data flow is divided into data frames with a length of 10 ms, ie 316 information bits are allocated to each frame. For residual error detection, these 316 bits are CRC-coded (CRC = Cyclic Redundancy Coding) and additionally provided with 16 bits redundancy. These 332 bits are now additionally appended with 8 tail bits in order to reset the convolutional encoder to a zero state after coding a data block (reset) and thus, if possible to allow reliable operation of a Viterbi decoder 20 'on the receiver side. With the help of the convolutional code (rate 1/3) these 340 bits are coded to 1020 bits, ie the data rate of the coded bits is 102 kbit / s. This rate is far from the next allowed channel data rates (64 kbps or 128 kbps).

A rate adjustment to the closest rate, i.e. to a channel data rate of 128 kbit / s (1280 bits per 10 ms) is therefore necessary. For this purpose, 260 of the 1020 bits are now repeated in order to fill up the 10 ms long frames to 1280 bits.

The trivial solution in which the received bit values, which correspond to the 260 repeated bits, would not be rejected without further evaluation is now not selected in the receiver E. Instead, the received bit values of repeated bits are compared with the received bit values at the original positions, and the value whose absolute value is the smaller one is discarded. For example, assuming that each bit is ideally received as software +1.0 or -1.0, a software value of 0.3 is received for a specific bit and a software value of 1.7 is received for the corresponding repetition bit , the value of 0.3 is rejected in this embodiment of the method according to the invention. The software value of 1.7 is then passed on to the channel decoder 20 'accordingly.

1020 received bit values are then forwarded to the subsequent Viterbi decoder 20 '. This data block is thus composed of 1020-260 = 760 receive bit values, which were not repeated in the transmitter, and the remaining 260 receive bit values, which were selected according to the method according to this embodiment.

FIG. 2 is a graphic representation of the frame error rate or bit error rate in the method according to the invention and in the trivial method.

2, the frame error rate or

Bit error rate is shown and on the y-axis the bit energy Eb normalized to the noise power density NO.

Curve A with the open triangles denotes the fra error rate in the trivial case, curve B with the open squares denotes the frame error rate in the case according to the invention, curve C with the closed triangles denotes the bit error rate Rate in the trivial case and curve D with the closed squares denotes the bit error rate in the case according to the invention.

The significant improvement in the respective error rate when using the method according to the invention can be clearly seen in FIG. 2 even with low normalized bit energy.

Although the present invention has been described above on the basis of a preferred exemplary embodiment, it is not limited to this, but can be modified in a variety of ways.

In particular, more than one repetition bit can be provided for certain bit values.

The invention is also not restricted to UMTS systems. Furthermore, coding / decoding or scrambling / descrambling are not absolutely necessary.

Claims

1. Data transmission method in which digital information signal is transmitted over a physical channel (K) between a transmitter (S) and a receiver (E), with the steps:

Providing the digital information signal in the form of at least one first bit data block with a first bit data rate in the transmitter (S);

Forming a second bit data block (B1, B2) from the bits of the first bit data block and one or more repeat bits with a second bit data rate in the transmitter (S);

Transmitting the second bit data block (B1, B2) over the physical channel (K) at the second bit data rate from the transmitter (S) to the receiver (E);

Forming a third bit data block from the bits of the transmitted second bit data block (B1 ', B2') with the second bit data rate in the receiver (E); and Forming a fourth bit data block from the bits of the third bit data block with the first bit data rate, wherein in the case of a group of receive bit values consisting of one bit and one or more corresponding repeat bits, a decision is made according to a predetermined algorithm, including at least two receive bit values, as to which bit value in the fourth bit data block is adopted.

2. The method according to claim 1, characterized in that certain pairs of bits and repetition bits are transmitted and, in the case of a pair consisting of a bit and a corresponding repetition bit, the maximum value of the amount of both receive bit values is determined and this is adopted in the fourth bit data block , while the smaller receive bit value is rejected.

3. The method according to claim 1, characterized in that certain pairs of bits and repetition bits are transmitted and in the case of a pair consisting of a bit and a corresponding repetition bit, the arithmetic mean of both received bit values is determined and this is adopted in the fourth bit data block.

4. The method according to claim 1, characterized in that certain pairs of bits and repeat bits are transmitted and, in the case of a pair consisting of a bit and a corresponding repeat bit, first of all root mean square

of both received bit values is determined in order to determine the amount of the value to be passed on to the fourth bit data block and that the sign of the value to be transferred to the fourth bit data block is chosen as the sign of the largest received bit value.

5. The method according to claim 1, characterized in that certain pairs of bits and repetition bits are transmitted and in the case of a pair consisting of a bit and a corresponding repetition bit, a first

Average value of any order w (a ^N + b ^N ) / N of both receivers

bit values is determined, where | α | and

6. The method according to claim 1, characterized in that certain pairs of bits and repetition bits are transmitted and in the case of a pair consisting of a bit and a corresponding repetition bit, a first

Average value of any order ^ a ^ ± \ b \ ') / N is determined, with l and \ b \ the amount of the respective received bit value denote, a is the receive bit value greater in amount and N = 3, 4, 5, ..., the amount of the in the fourth

Bit data block value to be adopted is determined to be W (α + \ b \) / N if the signs of a and b are identical

, and the amount too

is determined when the

Signs of the two received bit values are different, and that the sign of the value to be transferred to the fourth bit data block is selected in accordance with the sign of the largest received bit value a in terms of amount.

7. The method according to any one of the preceding claims, characterized in that the bits of the first bit data block are encoded before the formation of the second bit data block and the bits of the fourth bit data block are decoded accordingly.

8. The method according to any one of the preceding claims, characterized in that the bits of the first bit data block are made available before the formation of the second bit data block and the bits of the third bit data block are descrambled accordingly.

9. The method according to any one of the preceding claims, characterized in that the transmission is carried out in analog form, preferably via a radio link.

10. Data transmission method in which digital information signal is transmitted over a physical channel (K) between a transmitter (S) and a receiver (E), with:

an information signal input device (10) for providing the digital information signal in the form of at least one first bit data block with a first bit data rate in the transmitter (S);

rate adjustment means (40) for forming a second bit data block (B1, B2) from the bits of the first bit data block and one or more repeat bits with a second bit data rate in the transmitter (S);

a transmission device (50) for transmitting the second bit data block (B1, B2) over the physical channel (K) at the second bit data rate from the transmitter (S) to the receiver (E);

a rate adjustment device (40 ') for forming a third bit data block from the bits of the transmitted second bit data block (Bl', B2 ') with the second bit data rate in the receiver (E) and for forming a fourth bit data block from the bits of the third bit data block with the first bit data rate, wherein in the case of a group of receive bits consisting of one bit and one or more corresponding repeat bits, according to a predetermined algorithm, including at least two receive bits bits is decided which bit value is taken over into the fourth bit data block.