CN106788893B - Sparse interleaving multiple access method - Google Patents

Abstract

A sparse interleaving multiple access method proposed by the invention belongs to the technical field of non-orthogonal multiple access mobile communication. The method firstly makes all access users occupy K resource elements and T time-domain symbols when transmitting data blocks, and then performs channel coding on the data block information transmitted by each user, and divides the encoded data into several sub-bit streams , and map it into a codeword vector; zero-fill each codeword vector to obtain a K-dimensional sparse codeword vector; according to the sparse time-frequency interleaver, a K-dimensional sparse transmission codeword vector is obtained; The outgoing transmission resources transmit the corresponding components in the transmitted codeword vector; the receiver estimates the active users and their channel gains, and obtains their sparse time-frequency interleaver information; the receiver decodes to obtain the transmitted data blocks of each access user. The method enables each user to occupy only a small part of the total time-frequency resources when accessing and transmitting, thereby reducing the multi-user detection complexity of the receiver in the scenario of large-scale access.

Description

A sparse interleaving multiple access method

technical field

The invention relates to the technical field of non-orthogonal multiple access mobile communication, and specifically proposes a sparse interleave division multiple access method (Sparse Interleave Division Multiple Access, SIDMA).

Background technique

Due to the problems of multiple access interference (MAI) and inter-symbol interference (ISI) in the traditional code division multiple access (CDMA) method, its performance is limited. The existing main research focus is to propose a new multi-user detection method (multi-user detection, MUD) to reduce the impact of MAI and ISI on performance. In several documents, it is proposed to improve performance by assigning different interleavers to different users, so there is a traditional interleave division multiple access (IDMA) method. It has the advantages of traditional CDMA, that is, it resists fading through diversity, and alleviates the user interference problem of other cells in the worst case. Furthermore, due to the different random interleaving used by different users, adjacent chips are approximately uncorrelated, allowing the receiver to perform a relatively simple chip-by-chip iterative MUD technique. Through the soft information exchange iteration between the elementary signal estimator (Elementary Signal Estimator, ESE) and each user's respective posterior probability decoder (Decoder, DEC), the system can obtain performance close to Shannon limit. However, the traditional IDMA method requires users to occupy all the available resources to send data. In the case of large-scale access, this will lead to serious collision of orthogonal resources, which makes multi-user detection (MUD) extremely complicated.

In recent years, there have also been studies on the combination of IDMA and multi-antenna technology (Multi-Input Multi-Output, MIMO) to further improve the capacity and reception performance of the system. However, in general, in the case of large-scale user access , the existing methods generally face the problem of high computational complexity.

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide a sparse interleave division multiple access method (Sparse Interleave Division Multiple Access, SIDMA) in order to overcome the shortcomings of the prior art. The method of the invention enables each user to occupy only a small part of the total time-frequency resources when accessing and transmitting, thereby reducing the multi-user detection complexity of the receiver in the scenario of large-scale access.

A sparse interleaving multiple access method proposed by the present invention is characterized in that, the method firstly makes all access users occupy a total of K resource particles and T time-domain symbols when transmitting data blocks, and then compose the data transmitted by each user. The data block information is channel encoded, the encoded data is divided into several sub-bit streams, and mapped into codeword vectors; each codeword vector is filled with zeros to obtain a K-dimensional sparse codeword vector; according to the sparse time-frequency interleaver, we get K-dimensional sparse transmission codeword vector; at each moment, each user transmits the corresponding component in the transmission codeword vector according to the transmission resources given by the system at the corresponding moment; the receiver estimates the active users and their channel gains, and obtains their sparse time-frequency Interleaver information; the receiver decodes the received signal to obtain the transmitted data blocks of each access user. The method includes the following steps:

(1) All access users are allowed to occupy T time-domain symbols in total when transmitting data blocks, including K resource particles RE;

个比特；将向量

按每log₂M个比特分为一组，其中M为4的整数倍，表示调制阶数；根据码本，将向量

映射成长度为

的码字向量

N_i＜K；其中，码字向量

的每个分量通过1个RE进行传输；(2) Any access user u _i obtains an information bit vector after channel coding the information of the data block to be transmitted for this access Its length is

bits; convert the vector

Divide into a group by every log ₂ M bits, where M is an integer multiple of 4, indicating the modulation order; according to the codebook, the vector

Mapped to a length of

codeword vector of

N _i <K; where, the codeword vector

each component of Transmission through 1 RE;

进行补零，将原来的N_i维码字向量补零成为K维稀疏码字向量

其中K为RE的总个数；(3) Access the codeword vector obtained in step (2) by user u _i

Perform zero-filling, and zero-fill the original N _i -dimensional codeword vector into a K-dimensional sparse codeword vector

Where K is the total number of REs;

对步骤(3)得到的K维稀疏码字向量

进行交织映射，得到一个K维的稀疏发送码字向量：(4) Access user u _i according to the sparse time-frequency interleaver

For the K-dimensional sparse codeword vector obtained in step (3)

Perform interleaving mapping to obtain a K-dimensional sparse transmitted codeword vector:

Wherein the N _i positions are non-zero, that is, the N _i components in the corresponding step (2);

中的W(t)个分量，其中W(t)为时刻t的子载波数，并且按照系统的预设满足

(5) For any time t in T symbols, t=1, 2, ..., T, the access user u _i transmits its transmitted codeword vector according to the transmission resources given by the system at the corresponding time

W(t) components in , where W(t) is the number of subcarriers at time t, and meets the system preset

(6) The receiver identifies active users from all potential users registered on the network, that is, the users accessing this time, performs channel estimation, and obtains the sparse time-frequency interleaver corresponding to each active user according to the user identification results

(7) At each time t, t=1, 2, ..., T, the user access situation at the current time is obtained according to the identification result of step (6), and the receiver decodes the received signal y(t), where y (t) is the vector of dimension W(t), which represents the signal obtained by multiplying the W(t) components of the transmitted codeword vector received by the receiving end and sent by the access user at the corresponding time t by the channel gain of the corresponding user. ; Combine the decoding results of y(t) obtained at all times to perform channel decoding, and finally obtain the transmitted data blocks of each access user.

The characteristics and beneficial effects of the present invention are:

The present invention proposes an improved sparse interleaving multiple access (SIDMA) method after traditional IDMA. By introducing the sparseness of resource occupation, additional degrees of freedom are added. When there are few users, the probability of user collision at the resource is reduced, and the decoding complexity is low; when there are many users, collision can be tolerated; compared with traditional IDMA requires the consistency of the data rate of all users. SIDMA allows the amount of data transmitted by each user to be variable, that is, it can dynamically occupy resources according to the actual amount of data that needs to be transmitted, support variable rates, and be more flexible.

Description of drawings

Fig. 1 is the implementation flow chart of the method of the present invention.

FIG. 2 is a diagram illustrating an example of data block RE allocation in an embodiment of the present invention.

FIG. 3 is an example diagram of a sparse time-frequency interleaver for different users under a given RE in an embodiment of the present invention.

Detailed ways

A sparse interleaving multiple access method proposed by the present invention will be described in more detail below with reference to the accompanying drawings and specific embodiments.

A sparse interleaving multiple access method proposed by the present invention, as shown in Figure 1, includes the following steps:

(1) All access users are allowed to occupy a total of T time-domain symbols (symbols) when transmitting data blocks, including K resource elements (RE, Resource Element). 2 is an example diagram of a data block RE allocation situation, and an example of an RE allocation situation in which different time systems provide different numbers of subcarriers is given. In the figure, each square represents one RE, and there are k REs in total , all the squares as a whole represent the total available resources of randomly accessing all users at one time; each column of squares occupies one symbol time in the time domain, and occupies T symbols in total. The number of squares in each column represents the number of subcarriers available at the corresponding symbol time.

其长度为个比特，不同用户根据实际需求可传输不同长度的向量。将该向量

按每log₂M个比特分为一组，其中M(M＝4，8，16，…为4的整数倍)为调制阶数；根据码本，将该向量

映射成长度为

的码字向量其中，码字向量

的每个分量需要通过1个RE进行传输，因此要求N_i＜K，也即

码本的获取方式，可能包含但不限于，在随机接入的情况下，用户使用注册入网时分配的固定码本；在由调度接入的情况下，用户使用系统调度时分配的码本。(2) Any access user u _i obtains an information bit vector after channel coding the information of the data block to be transmitted for this access

Its length is Different users can transmit vectors of different lengths according to actual needs. the vector

Divide each log ₂ M bits into a group, where M (M=4, 8, 16, ... is an integer multiple of 4) is the modulation order; according to the codebook, the vector

map to a length of

codeword vector of Among them, the codeword vector

each component of It needs to be transmitted through 1 RE, so it is required that _Ni <K, that is,

The acquisition method of the codebook may include, but is not limited to, in the case of random access, the user uses the fixed codebook allocated when registering to access the network; in the case of scheduling access, the user uses the codebook allocated during system scheduling.

进行补零，将原来的N_i维码字向量补零成为K维稀疏码字向量

其中K为RE的总个数。(3) Access the codeword vector obtained in step (2) by user u _i

Perform zero-filling, and zero-fill the original N _i -dimensional codeword vector into a K-dimensional sparse codeword vector

where K is the total number of REs.

对步骤(3)得到的K维稀疏码字向量

进行交织映射，得到一个K维的稀疏发送码字向量：(4) Access user u _i according to the sparse time-frequency interleaver

For the K-dimensional sparse codeword vector obtained in step (3)

Perform interleaving mapping to obtain a K-dimensional sparse transmitted codeword vector:

中的N_i个分量。稀疏时频交织器的获取方式，可能包含但不限于，在随机接入的情况下，用户使用注册入网时分配的固定稀疏时频交织器；在由调度接入的情况下，用户使用系统调度时分配的稀疏时频交织器。Among them, only N _i positions are non-zero, that is to say, they carry the steps in step (2).

N _i components in . The acquisition method of the sparse time-frequency interleaver may include, but is not limited to, in the case of random access, the user uses the fixed sparse time-frequency interleaver allocated when registering to the network; in the case of scheduling access, the user uses the system scheduling Time-allocated sparse time-frequency interleaver.

的信息比特，采用M＝16-QAM调制，它们的码字向量长度分别为N₁＝N₂＝N₃＝10。在给定的K＝40个传输资源RE下，进行补零后得到的40维稀疏码字向量，也即每个稀疏时频交织器的输入是：Figure 3 is an example diagram of a sparse time-frequency interleaver for different users with a given resource RE, and Figure 3 shows possible three users u ₁ , u ₂ , u ₃ given K=40 REs Below, each person sends a length of

information bits, using M=16-QAM modulation, their codeword vector The lengths are respectively N ₁ =N ₂ =N ₃ =10. Under the given K=40 transmission resources RE, the 40-dimensional sparse codeword vector obtained after zero-filling, that is, the input of each sparse time-frequency interleaver is:

i＝1，2，3，除了阴影对应的10个位置包含了输入的10个码字外，其余位置均为0。定义

表示用户发送码字的稀疏程度。The output is a sparse 40-dimensional transmitted codeword vector as shown in Figure 3

i=1, 2, 3, except that the 10 positions corresponding to the shadow contain the 10 input codewords, the other positions are all 0. definition

Indicates the sparsity of the codeword sent by the user.

(5) For any time t in T symbols, t=1, 2, ..., T, the access user u _i transmits its transmitted codeword vector according to the transmission resources given by the system at the corresponding time W(t) components in , where W(t) is the number of subcarriers at time t, and meets the system preset

也即每个时刻系统均分配有4个RE。用户u₁，u₂，u₃在对应时刻t发送图2所示的第t列的长度为4的列向量即可。For the case shown in Figure 3, W(t)=4, t=1, 2, . . . , 10, and

That is, 4 REs are allocated to the system at each moment. Users u ₁ , u ₂ , and u ₃ may send a column vector with a length of 4 in the t-th column shown in FIG. 2 at the corresponding time t.

(6) The receiver identifies the active user (that is, the user accessing this time) from all potential users registered in the network, performs channel estimation, and obtains the sparse time-frequency interleaver corresponding to each active user according to the user identification result.

(7) At each time t, t=1, 2, ..., T, the user access situation at the current time is obtained according to the identification result of step (6), and the receiver decodes the received signal y(t), where y (t) is the vector of dimension W(t), which indicates that the W(t) components of the transmitted codeword vector received by the receiving end and sent by the access user at the corresponding time t are multiplied by the channel gain of the corresponding user and then superimposed. signal, and may also contain noise depending on the transmission environment. Finally, the decoding results of y(t), t=1, 2, .

The decoding method adopted in this embodiment may, but is not limited to, use a message passing algorithm (Message Passing Algorithm, MPA).

Claims (2)

1. A sparse interleaving multiple access method is characterized in that the method firstly makes all access users occupy K resource particles and T time domain symbols when transmitting data blocks, then carries out channel coding on the data block information transmitted by each user, divides the coded data into a plurality of sub bit streams and maps the sub bit streams into code word vectors; zero padding is carried out on each code word vector to obtain a K-dimensional sparse code word vector; obtaining a K-dimensional sparse sending codeword vector according to a sparse time-frequency interleaver; at each moment, each user transmits and sends a corresponding component in the code word vector according to the transmission resource given by the system at the corresponding moment; the receiver estimates active users and channel gains thereof, and acquires sparse time-frequency interleaver information thereof; and the receiver decodes the received signals to obtain the transmission data blocks of all the access users.

2. A method according to claim 1, characterized in that the method comprises the steps of:

(1) the method comprises the steps that T time domain symbols symbol are allowed to be occupied totally when all access users transmit data blocks, and K resource elements RE are contained;

(2) any access user u_iChannel coding the information of the data block to be transmitted in the access to obtain an information bit vector

Having a length ofA bit; will vector

Per log₂M bits are divided into a group, wherein M is an integral multiple of 4 and represents a modulation order; according to the codebook, vectors are encoded

Is mapped to a length ofCode word vector of

N_i< K; wherein the code word vector

Each component of

Transmitting through 1 RE;

(3) access user u_iFor the code word vector obtained in the step (2)

Zero filling is carried out to fill the original N_iZero-filling of dimensional codeword vectors into K-dimensional sparse codeword vectors

Wherein K is the total number of RE;

(4) access user u_iTime-frequency interleaver according to sparseness

For the K-dimensional sparse code word vector obtained in the step (3)

Performing interleaving mapping to obtain a K-dimensional sparse sending code word vector:

wherein N is_iOne position being non-zero, i.e. corresponding to N in step (2)_iA component;

(5) for any time T, T equals 1,2, …, T in T symbols, access user u_iTransmitting the sending code word vector according to the transmission resource given by the corresponding time systemW (t) components, where w (t) is the number of subcarriers at time t, and is satisfied according to a preset of the system

(6) The receiver identifies active users, namely the users accessed this time, from all potential users registered in the network, performs channel estimation, and obtains a sparse time-frequency interleaver corresponding to each active user according to the user identification result

(7) At each time T, T is 1,2, …, T, the user access situation at the current time is obtained according to the identification result of step (6), and the receiver decodes the received signal y (T), wherein y (T) is a vector with the length w (T) and represents a signal obtained by multiplying the w (T) components of the transmitted code word vector transmitted by the access user at the corresponding time T received by the receiving end by the channel gain of the corresponding user and then superimposing; and (5) jointly performing channel decoding on the decoding results of y (t) obtained at all the moments to finally obtain the sending data block of each access user.

Images