In a wireless broadcast system, the quality of wireless links of users located in different locations is very different, the channel conditions between some users and the transmitting station are relatively good, while others are very bad. In order to guarantee the quality of service of all users in the service area, the system usually designs the transmission parameters according to the conditions that can guarantee a certain quality of service under the worst conditions, which makes many users with better channel conditions only have poor channel conditions. Users get the same rate of data, resulting in a lower overall system transfer rate. In addition, there will be various types of receiving devices in future broadcast systems, their usage conditions are different, and the receiving capabilities are also very different. For example, a high quality indoor receiver should be able to receive high rate data, while a small portable device only needs to receive lower rate data. In order to make full use of the transmission capacity of the channel and meet the requirements of different users and different receiving devices, it is necessary to design a system with multi-level transmission. The concept of multi-level transmission was first proposed by Cover [1]. He pointed out that the hierarchical transmission scheme can meet the needs of users with different channel conditions and obtain higher transmission efficiency. A two-stage coded modulation scheme is proposed in the European Digital Television Broadcasting Standard (DVB-T) [2]. Literature [3] discusses the different error protection characteristics brought by multi-level code modulation and multi-level decoding. The literature [4] analyzes the performance of multi-level modulation scheme in uncoded systems. In recent years, with the development of communication technology, efficient codec methods have emerged, providing conditions for designing better broadcast transmission schemes. This paper proposes a multi-level transmission scheme for digital radio broadcasting systems, which uses low-density parity check code (LDPC) to hierarchically encode broadcast data, and combines multi-level coding and multi-resolution modulation techniques. The priority data provides different error protection, so that each user can obtain different data transmission rates according to the quality of the transmission channel or the capability of the receiving device, thereby improving the overall transmission efficiency of the system.
This article refers to the address: http://
Multi-level transmission system
The multi-level transmission system refers to dividing the data to be transmitted into different priorities according to different importance or content, and separately processing the data of each priority in the coding and modulation process of the transmitting end, so that the data of different priorities are obtained differently. The error protection, the decoding threshold of the high priority data is lower, the decoding is relatively easy, the decoding threshold of the low priority data is higher, and the decoding is relatively difficult. At the receiving end, the user selects to demodulate only high priority data or demodulate all data according to the received signal quality or the processing capability of the receiving device. This not only ensures the basic coverage of the network, but also enables users with better channel conditions or strong receiving capabilities to obtain more data, thereby increasing the average transmission rate of the system.
The multi-level transmission system mainly consists of three parts: code modulation, channel transmission, and demodulation and decoding. As shown in Figure 1, the block diagram of the multi-stage transmission system, the data to be transmitted is divided into parallel multi-channel data streams according to different priorities, and enters the multi-level encoder. Different encoders independently encode each channel of data, and the code symbols after encoding are first subjected to set partitioning and constellation mapping, and then OFDM modulated into the wireless channel for transmission. At the receiving end, channel estimation is performed according to the received signal, and then demodulation and hierarchical decoding are performed to obtain data of respective priorities.
Figure 1 multi-level transmission system
2.1 multi-level coding
To achieve multi-level transmission, the data must first be multi-level encoded. Different encoders separately encode the data of each priority, which is called a component code. The component code can conveniently adjust the encoding rate of each priority data to provide different error protection for each channel of data. Currently, the component code coding methods commonly used are RS code, convolutional code, Turbo code, LDPC code, and the like.
The LDPC code was first proposed by Gallager in 1962 and was rediscovered and promoted at the end of the 20th century. The LDPC code has a powerful error correction capability to achieve performance close to the Shannon limit. The LDPC code is a linear block code. The super-sparse random matrix is ​​used as the check matrix. The check nodes and the bit nodes are constrained by the row and column of the check matrix. The main decoding method is confidence. Propagation algorithm. This scheme adopts the LDPC code [5] in the digital terrestrial broadcasting standard of China. The code rate is 0.4, 0.6, and 0.8. The data of each level is encoded by LDPC with different code rates, which can make different priorities have different data. The degree of protection meets the needs of hierarchical transmission systems.
2.2 set partitioning and constellation mapping
The mapping relationship between the coded symbol symbols and the signal points is also called signal set segmentation [6]. The selection of the set partitioning scheme is the key to designing the coding and modulation. The channel coding and modulation methods can be combined by the set division. This scheme selects a set segmentation form suitable for hierarchical transmission. This segmentation method divides the points on the constellation diagram into different clusters, and each cluster is divided into several sub-clusters. The distance between adjacent clusters is greater than the distance between adjacent sub-clusters, the high-priority data is mapped onto the cluster, and the second-priority data is mapped to the sub-cluster. This multi-level segmentation method can provide different data. Level error protection.
Figure 264QAM constellation mapping
Figure 2 shows a 64-QAM constellation map [7], where d1, d2, and d3 determine the distance between each signal point. In the figure, each constellation point represents 6-bit data (represented by X0X1X2X3X4X5, respectively).
Figure 3 set split
Figure 3 shows the specific process of set partitioning. According to the first two bits (X0 and X1), the constellation can be divided into four quadrants with 16 points in each quadrant. According to the difference between the two bits (X2 and X3), 16 points in each quadrant can be subdivided into 4 parts. Finally, each signal point is distinguished according to the following two bits (X4 and X5). The data thus transmitted is divided into three levels, X0 and X1 are high priority, X2 and X3 are medium priority, and X4 and X5 are low priority. The demodulation complexity of high-priority data is equivalent to QPSK, and the difficulty of demodulation is relatively low, that is, the performance is better at a lower transmission rate; and for the medium and low priority data, it is equivalent to 16QAM and 64QAM demodulation, so that the performance of different priority data has a clear distinction, meeting the needs of hierarchical transmission.
Since the coding of the component code, the set division and the constellation mapping can provide different error protection for each priority data, it is necessary to combine the coding and modulation methods to uniformly design and allocate the parameters of each part, which is better. The ground achieves the purpose of hierarchical transmission.
2.3 channel model
In wireless broadcast transmission, the signal will be reflected by the influence of mountains, buildings, moving objects, so that signals arriving at the receiver through different paths will have obvious multipath effects, resulting in signal fading. In addition, in order to improve spectrum utilization, a single-frequency network networking method will be adopted. The so-called single frequency network [8] means that each transmitting station transmits the same data synchronously using the same frequency, which is advantageous for mobile reception and frequency planning. However, it is technically necessary to overcome the special "multi-source multipath" problem, that is, to process complex multipath signals arriving from different transmitting stations, in different directions, and at different times.
According to the channel characteristics described above, the channel model in the simulation adopts a phase modulation fading model, and the multipath fading channel can be expressed as
Among them, hl, Ï„l, and fl represent the amplitude, delay, and Doppler shift of different paths, respectively. In this model, it is assumed that hl, Ï„l, and fl of each path are not related to each other.
2.4 Channel Estimation
Due to the large multipath effect in the wireless broadcast transmission, especially in the case of a single frequency network, the influence of multipath is more severe, resulting in difficulty in reception, and therefore a higher requirement is imposed on the method of channel estimation. Only receivers can make a more accurate estimate of the channel to get better reception performance. Referring to the Chinese digital television terrestrial transmission standard, the channel estimation is performed by using the PN sequence inserted in front of the signal frame. The calculation complexity is small, and the characteristics of the channel can be quickly obtained.
Assuming that the length of the PN sequence is K, according to the autocorrelation of the m-sequence, the normalized cyclic autocorrelation function is
Therefore, the channel response function can be obtained by performing a sliding correlation operation with the locally generated PN sequence and the received sample:
2.5 hierarchical demapping and multi-level decoding
At the receiving end, in order to obtain better performance, constellation demapping uses a soft decision method. The soft output at the 64QAM constellation map mapping is the log likelihood ratio of each bit of data. According to the literature [9]:
Where r is the received signal, C is the channel state information, and y is the signal y=r/C after equalization. For 64QAM, the log likelihood ratio of each priority bit can be approximated as [9]
Among them, b2, b1 and b0 respectively represent data of three priority levels of high, medium and low, and the soft decision output of each level of data can be obtained separately. This method of soft decision is not only simple to operate, but also capable of achieving good performance.
The soft output data obtained by the demapping enters the multi-stage decoder for decoding, and the multi-level decoder can separately decode the data of each priority separately, so the receiver can terminate the decoding process at any time after obtaining the required data. . For small portable receivers, due to screen size, battery capacity and processing capacity limitations, only high-priority data can be decoded to meet the requirements, thus reducing the computational complexity and complexity of the decoding circuit, and reducing Device power consumption is better adapted to the needs of portable devices. For larger fixed receivers, the screen is large and the processing power is strong. When the channel conditions are good, all priority data can be decoded to provide the highest quality service. When the channel conditions are poor, the low priority data Serious errors will occur, and high-priority data will still be correctly decoded due to the lower signal-to-noise ratio threshold required for decoding, ensuring the most basic quality of service. In addition, when each priority code stream transmits a different program, the decoder can decode only a certain priority code stream to obtain corresponding data without decoding other priority code streams, thereby reducing reception. Machine power consumption. Therefore, the method using multi-level decoding can better meet the requirements of different types of devices and different channel conditions, and has great flexibility.
Simulation results and analysis
In the simulation system, the data to be transmitted is divided into three priority levels, and each priority data is encoded by an LDPC having a code length of 7488 and a code rate of 0.4, 0.6 or 0.8. The constellation map uses the 64QAM constellation diagram shown in Figure 2. The signal point spacing is d1:d2:d3=1:1:1. The data obtained after mapping is subjected to symbol interleaving, frequency domain interleaving and 3780 carrier OFDM modulation. . The channel models used for the simulation are AWGN channels and multipath channels. At the receiving end, the PN sequence of length 255 is first used for channel estimation, and then the demapping and multi-level decoding methods described above are used to independently decode the high, medium and low priority data, and calculate separately. The bit error rate of each priority data.
Table 1 Decoding performance of each priority data in the AWGN channel
Table 1 shows the performance of each LDPC code with different bit rates for each priority data in the AWGN channel. In the simulation, the decoding threshold is calculated according to the bit error rate less than 3×10−6. It can be seen from the simulation results that the combination of different priority data and different code rates can obtain multiple decoding thresholds, and the threshold of each priority data can be flexibly selected according to different situations to meet different types of requirements. Furthermore, the thresholds obtained even when each priority data uses LDPC encoding of the same code rate are different, thereby verifying that different bits of the constellation point have different protection characteristics.
4 is a performance comparison curve of a hierarchical transmission scheme under the AWGN channel and a conventional non-hierarchical transmission scheme. The high, medium and low priority data of the hierarchical transmission are respectively coded by using LDPC codes with a code rate of 0.4, 0.6, and 0.8 to form 3 code streams and 3 decoding thresholds. In the case of non-hierarchical transmission, all data is uniformly encoded by using an LDPC code with a code rate of 0.6 to form a single code stream and a decoding threshold. The total transmission rates of the two schemes are equal. It can be seen from the curve that in the hierarchical transmission mode, the threshold gap between the high, medium and low priority data is 8.4 dB and 5.4 dB, and the priority data required for the hierarchical transmission has a clear performance differentiation target.
Figure 4 Comparison of performance between hierarchical and non-hierarchical transmissions in the AWGN channel
The high- and medium-priority data performance of the hierarchical mode is better than the non-hierarchical mode, so that the correct transmission of some data can still be guaranteed at a low SNR, and the decoding does not completely fail like the non-hierarchical mode, thereby improving the average of the system. Transmission rate. At the same time, the hierarchical transmission mode increases the decoding threshold of low priority data relative to the non-hierarchical mode, but in consideration of the need to receive low priority data, generally a fixed receiver, the performance loss can be compensated by using a high gain directional antenna. In addition, the method of reducing the low priority data rate can also be used to improve the performance, at the cost of losing a certain transmission rate.
Table 2 Decoding performance of each priority data under multipath channel
Table 2 shows the performance of the high, medium, and low priority data in the hierarchical mode of the multipath channel using LDPC codes of 0.4, 0.6, and 0.8, respectively, and the LDPC code using a single 0.6 in the non-hierarchical mode. The channel model uses the Brazilian multipath model. For specific channel parameters, see [10].
Figure 5 Comparison of performance of hierarchical and non-hierarchical transmission under multipath channel
Fig. 5 is a performance comparison curve of hierarchical transmission and non-hierarchical transmission under multipath channel. The parameters used in the simulation are the same as those in Fig. 4, and the channel adopts the Brazilian B model. From the results in the graphs and tables, it can be seen that the performance of each priority data in the multipath channel has a different degree of decline, but the threshold difference required for the hierarchical transmission is still maintained, and the performance of the high priority data is minimized. This ensures that high priority data can still be decoded correctly under harsh channel conditions.
Figure 6 Comparison of the performance of the 3-level transmission scheme and the 2-level transmission scheme
Figure 6 shows the performance comparison between the three-level transmission scheme proposed in this paper and the two-level transmission scheme in the DVB-T standard under the AWGN channel. In the simulation, the three priority data in the scheme use 0.4, 0.6, and 0.8 code rates respectively. The LDPC code, while the two priority data in the DVB-T scheme uses punctured convolutional codes of 1/2 and 5/6 code rate, so that the total transmission rates of the two schemes are basically the same. It can be seen from the figure that the decoding thresholds of the three priority data of the scheme are 4.8dB, 13.2dB and 18.6dB, respectively, and the decoding thresholds of the two priority data of the DVB-T scheme are 6.5dB and 21.9dB, The priority data performance of the solution is significantly better than the DVB-T solution.
This paper proposes a multi-level transmission scheme for digital radio broadcasting system, which divides broadcast data into multiple priorities, uses LDPC codes as component codes for multi-level coding, and uses hierarchical set segmentation and constellation mapping strategies. Provide different error protection for different priority data. The method of hierarchical demapping and multi-level decoding is adopted at the receiving end, so that users of different channel conditions and receiving devices can obtain different transmission rates, and the purpose of hierarchical transmission of broadcast data is achieved. The simulation results show that the proposed scheme outperforms the traditional unclassified scheme and the DVB-T grading scheme under the AWGN channel and various multipath channels, and also improves the average transmission rate of the system.
Electric Bicycle Battery,24V Ebike Battery Series,24V Cylinder Lithium Battery,Deep Cycle Lithium Battery
ZHEJIANG TIANHONG LITHIUM-ION BATTERY CO.,LTD , https://www.tflbattery.com