CN107359925A - A kind of virtual full duplex relaying transmission method - Google Patents

Abstract

The present invention relates to a virtual full-duplex relay transmission method. In an odd time slot 2t-1, K source terminals S ₁ , S ₂ , S ₃ ... S _K send signals to the relay R ₁ , at this time The relay R ₂ forwards the signal received in the previous time slot 2t‑2 to K sink terminals D ₁ , D ₂ ,...D _K after decoding and re-encoding. In the even time slot 2t, K sources Terminals S ₁ , S ₂ , S ₃ ... S _K send signals to relay R ₂ , and relay R ₁ decodes and re-encodes the signal received in the previous time slot 2t‑1 and forwards it to K sink terminals D ₁ , D ₂ ,...D _K , relay R ₁ and relay R ₂ take turns to receive and send signals, K source terminals S ₁ , S ₂ , S ₃ ...S _K always send signals, The K sink terminals D ₁ , D ₂ , ... D _K are always receiving signals, achieving the same transmission efficiency as the full-duplex relay method using two ordinary half-duplex The relay sends and receives signals alternately, thus avoiding the phenomenon of full-duplex self-interference.

Description

A virtual full-duplex relay transmission method

technical field

The invention relates to the technical field of wireless communication, in particular to a virtual full-duplex relay transmission method.

Background technique

In a wireless communication system, when the distance between the transmitting end and the receiving end is long, the transmitting end needs to consume a large amount of power to maintain normal communication. By introducing a relay terminal between the transmitting end and the receiving end for signal forwarding, this can be done in a certain To reduce the terminal power consumption to a certain extent, this technology is called wireless relay technology. As shown in Figure 1, a typical wireless relay system is given, in which K (K≥1) source terminals S ₁ , S ₂ , S ₃ ... S _K will The signal is sent to corresponding _K sink terminals D ₁ , D ₂ , D ₃ .

For the above-mentioned wireless relay communication system, half-duplex relay transmission is usually adopted, that is, two time slots are used to complete an information transmission, as shown in Figure 1, in odd time slots, all source terminals send signals to the relay Terminal R; in even time slots, the relay terminal R forwards the signal to the destination terminal after certain processing. Since this method needs 2 time slots to complete one information transmission, the efficiency is low. In order to improve the transmission efficiency, a full-duplex relay transmission technology has been proposed in recent years, that is, the relay terminal R is a full-duplex terminal that can simultaneously receive and transmit signals. This full-duplex relay transmission is shown in Figure 2. The source terminal always sends signals to the relay R, and the relay R adopts the full-duplex mode. While receiving the signal from the source terminal, the signal received in the previous time slot is processed and then sent to the sink terminal, and the sink terminal Then the forwarding signal from the relay terminal is always received. Compared with the traditional half-duplex relay, this kind of full-duplex relay transmission effectively improves the transmission efficiency, but the full-duplex relay terminal also has certain disadvantages, that is, there are serious The phenomenon of self-interference, and the current existing technologies all adopt analog or digital interference cancellation technology, which is very complicated and causes high hardware cost and implementation complexity.

Contents of the invention

The technical problem to be solved by the present invention is to provide a virtual full-duplex relay transmission method that can ensure signal transmission efficiency and prevent self-interference in relay terminals.

The technical solution adopted in the present invention is a virtual full-duplex relay transmission method, which is completed through a signal transmission system, and the signal transmission system includes K source terminals S ₁ , S ₂ , S ₃ ... S _K , relay R ₁ , relay R ₂ and K sink terminals D ₁ , D ₂ , D ₃ ... D _K , the relay R ₁ and the relay R ₂ are deployed with M antennas, and each source The transmit power of the terminal is P _S , and the transmit power of relay R ₁ and relay R ₂ are P _R1 and P _R2 respectively. This transmission method uses 2 time slots to complete a signal transmission, and the odd time slot is set as 2t- 1. The even time slot is 2t, where t=1,2,3...n, the method includes the following steps:

(1) In the odd time slot 2t-1, K source terminals S ₁ , S ₂ , S ₃ ... S _K send signals to the relay R ₁ , at this time, the relay R ₂ transmits the previous time slot 2t -2 The received signal is forwarded to K sink terminals D ₁ , D ₂ ,...D _K after being decoded and re-encoded;

(2) In the even time slot 2t, K source terminals S ₁ , S ₂ , S ₃ ... S _K send signals to the relay R ₂ , and the relay R ₁ receives the signal from the previous time slot 2t-1 The signal of is forwarded to K sink terminals D ₁ , D ₂ ,...D _K after being decoded and re-encoded;

(3) Return to step (1) to continue sending and receiving signals until all signals are transmitted.

In step (1), the signal to be sent by a source terminal S _k is set as s _k (2t-1), k=1, 2, 3...K, then the transmitted signals of K source terminals are used The vector representation is: s(2t-1)=[s ₁ (2t-1),...s _K (2t-1)] ^T .

In step (1), the signals of K source terminals received and detected by relay R ₂ in the previous time slot (2t-2) can be expressed as: After decoding and re-encoding, it can be concluded that the signal sent by relay R ₂ to K sink terminals can be expressed as: where _α2 represents the power control coefficient of relay _R2 , namely Where tr( ) represents the trace operation, Represents the channel coefficient matrix from relay R ₂ to K sink terminals, its size is K×M, and the superscript (·) ^H represents the conjugate transpose.

In step (1), when relay R ₁ receives signals from K source terminals, at this time, relay R ₁ will receive superimposed signals from K source terminals and relay R ₂ , then The superimposed signal can be expressed as: Among them, H _SR1 represents the channel coefficient matrix from K source terminals to relay R ₁ , and its size is M×K, and G _R2R1 represents the channel coefficient matrix from relay R ₂ to relay R ₁ , and its size is M×M, z _R1 (2t- ₁ ) represents the white Gaussian noise of relay R1.

In step ( ₁ ), the signal received by the k-th destination terminal D _k from relay R2 can be expressed as: in Denotes the channel vector from relay R ₂ to the kth terminal D _k , its size is 1×M, z _D,k (2t-1) represents the Gaussian white noise of k source terminals D _k , k=1,2 ,3...K.

In step (2), set the signal to be sent by the source terminal S _k as s(2t), k=1, 2, 3...K, then the transmitted signals of the K source terminals can be represented by a vector It is: s(2t)=[s ₁ (2t),....s _K (2t)] ^T .

In step (2), the signals of K source terminals obtained by relay R ₁ receiving and detecting signals in the previous time slot (2t-1) can be expressed as: After decoding and re-encoding, it can be concluded that the signal sent by relay R ₁ to K sink terminals can be expressed as: α ₁ represents the power control coefficient of relay R ₁ , which can be expressed as Where tr( ) represents the trace operation, Represents the channel coefficient matrix of relay R ₁ to K sink terminals, its size is K×M, and the superscript (·) ^H represents the conjugate transpose.

In step (2), when the relay R ₂ receives signals from K source terminals, the relay R ₂ at this time will receive the superimposed signals from the K source terminals and the relay R ₁ , then The superimposed signal can be expressed as: where H _SR2 represents the channel coefficient matrix from K source terminals to relay R ₂ , and its size is M×K; G _R1R2 represents the channel coefficient matrix from relay R ₁ to relay R ₂ , and its size is M×M , z _R2 (2t) represents the white Gaussian noise of relay R ₂ .

In step (2), the signal received by the k-th destination terminal D _k from relay R ₁ can be expressed as: in Denotes the channel vector from relay R ₁ to the k-th source terminal D _k , its size is 1×M, z _D,k (2t) represents the Gaussian white noise of the k-th source terminal D _k .

The beneficial effect of the present invention is that: it can be seen from the above method that the relay R ₁ and the relay R ₂ take turns to receive and send signals, and the K source terminals S ₁ , S ₂ , S ₃ ... S _K always send signals, while the K sink terminals D ₁ , D ₂ ,...D _K are always receiving signals, achieving the same transmission efficiency as the full-duplex relay method using two ordinary half The duplex relay sends and receives signals alternately, thus avoiding the phenomenon of full-duplex self-interference, and there is no need to design a dedicated full-duplex self-interference cancellation circuit, thereby reducing the hardware complexity of the communication system while maintaining the same performance as full-duplex rate performance.

Description of drawings

FIG. 1 is a schematic diagram of a traditional half-duplex relay transmission technology based on two time slots in the background technology;

FIG. 2 is a schematic diagram of full-duplex relay transmission technology in the background technology;

3 is a schematic diagram of a virtual full-duplex relay transmission method of the present invention;

Fig. 4 is the performance comparison figure that a kind of virtual full-duplex relay transmission method proposed by the present invention compares with other methods;

detailed description

The invention will be further described below with reference to the accompanying drawings and in combination with specific embodiments, so that those skilled in the art can implement it by referring to the description, and the protection scope of the present invention is not limited to the specific embodiments.

The present invention designs a virtual full-duplex relay transmission method, which is completed through a signal transmission system, as shown in Figure 3, the signal transmission system includes K source terminals S ₁ , S ₂ , S ₃ ... S _K , relay R ₁ , relay R ₂ and K sink terminals D ₁ , D ₂ , D ₃ ... D _K , this transmission method uses 2 time slots to complete a signal transmission, and the odd time slots are set as 2t-1, the even time slot is 2t, wherein t=1,2,3...n, the method includes the following steps:

(1) In the odd time slot 2t-1, K source terminals S ₁ , S ₂ , S ₃ ... S _K send signals to the relay R ₁ , and the relay R ₂ sends the current time slot The signal received in the previous time slot 2t-2 is decoded and re-encoded and forwarded to K sink terminals D ₁ , D ₂ ,...D _K ;

(2) In the even time slot 2t, K source terminals S ₁ , S ₂ , S ₃ ... S _K send signals to the relay R ₂ , and the relay R ₁ transmits the signal to the previous time slot 2t of the current time slot -1 The received signal is decoded and re-encoded and forwarded to K sink terminals D ₁ , D ₂ ,...D _K ;

(3) Return to step (1) to continue sending and receiving signals until all signals are transmitted.

Both the relay R ₁ and the relay R ₂ deploy M antennas, the transmission power of each source terminal is PS, the transmission powers of the relay R ₁ and the relay R ₂ are _{P R1} and P _R2 respectively, and the step ( In 1), the signal to be sent by a source terminal S _k is set as s _k (2t-1), k=1, 2, 3...K, then the transmitted signals of K source terminals are represented by vectors as : s(2t-1)=[s ₁ (2t-1),...s _K (2t-1)] ^T .

In step (1), the signals of K source terminals received and detected by relay R ₂ in the previous time slot (2t-2) can be expressed as: After decoding and re-encoding, it can be concluded that the signal sent by relay R ₂ to K sink terminals can be expressed as: where _α2 represents the power control coefficient of relay _R2 , namely Where tr( ) represents the trace operation, Represents the channel coefficient matrix from relay R ₂ to K sink terminals, its size is K×M, and the superscript (·) ^H represents the conjugate transpose.

In step (1), when relay R ₁ receives signals from K source terminals, at this time, relay R ₁ will receive superimposed signals from K source terminals and relay R ₂ , then The superimposed signal can be expressed as: Among them, H _SR1 represents the channel coefficient matrix from K source terminals to relay R ₁ , and its size is M×K, and G _R2R1 represents the channel coefficient matrix from relay R ₂ to relay R ₁ , and its size is M×M, z _R1 (2t- ₁ ) represents the white Gaussian noise of relay R1.

In step ( ₁ ), the signal received by the k-th destination terminal D _k from relay R2 can be expressed as: in Denotes the channel vector from relay R ₂ to the kth terminal D _k , its size is 1×M, z _D,k (2t-1) represents the Gaussian white noise of k source terminals D _k , k=1,2 ,3...K.

In step (2), the signal to be sent by the source terminal S _k is set as s _k (2t), k=1, 2, 3...K, then the transmitted signals of the K source terminals can be expressed by vector Expressed as: s(2t)=[s ₁ (2t),....s _K (2t)] ^T .

In step (2), the signals of K source terminals obtained by relay R ₁ receiving and detecting signals in the previous time slot (2t-1) can be expressed as: After decoding and re-encoding, it can be concluded that the signal sent by relay R ₁ to K sink terminals can be expressed as: α ₁ represents the power control coefficient of relay R ₁ , which can be expressed as Where tr( ) represents the trace operation, Represents the channel coefficient matrix of relay R ₁ to K sink terminals, its size is K×M, and the superscript (·) ^H represents the conjugate transpose.

In step (2), when the relay R ₂ receives signals from K source terminals, the relay R ₂ at this time will receive the superimposed signals from the K source terminals and the relay R ₁ , then The superimposed signal can be expressed as: where H _SR2 represents the channel coefficient matrix from K source terminals to relay R ₂ , and its size is M×K; G _R1R2 represents the channel coefficient matrix from relay R ₁ to relay R ₂ , and its size is M×M , z _R2 (2t) represents the white Gaussian noise of relay R ₂ .

In step (2), the signal received by the k-th destination terminal D _k from relay R ₁ can be expressed as: in Denotes the channel vector from relay R ₁ to the k-th source terminal D _k , its size is 1×M, z _D,k (2t) represents the Gaussian white noise of the k-th source terminal D _k .

As shown in Figure 4, using 10 source terminals and 10 sink terminals, the virtual full-duplex relay method proposed by the present invention in the figure has achieved the same rate performance as the full-duplex relay, which is better than the traditional half-duplex Relay method.

Claims (9)

1. a kind of virtual full duplex relaying transmission method, is completed, it is whole that the communication system includes K information source by communication system Hold S₁,S₂,S₃...S_K, relaying R₁, relaying R₂And K destination end D₁,D₂,D₃...D_K, the relaying R₁With relaying R₂Equal portion M root antennas are affixed one's name to, the transmit power of each information source terminal is P_S, relay R₁With relaying R₂Transmit power be respectively P_R1And P_R2, the biography Transmission method completes a signal transmission using 2 time slots, sets odd numbered slots as 2t-1, even timeslots 2t, wherein t=1, 2,3...n, methods described comprises the following steps：

(1), in odd numbered slots 2t-1, K information source terminal S₁,S₂,S₃...S_KSend a signal to relaying R₁, now relay R₂Then will The signal that previous time slot 2t-2 is received is transmitted to K destination end D after decoding and recompiling₁,D₂,...D_K；

(2), in even timeslots 2t, K information source terminal S₁,S₂,S₃...S_KSend a signal to relaying R₂, relay R₁When then will be previous The signal that gap 2t-1 is received is transmitted to K destination end D after decoding and recompiling₁,D₂,...D_K；

(3), return to step (1) continues the transmission and reception of signal, until all signal end of transmissions.

A kind of 2. virtual full duplex relaying transmission method according to claim 1, it is characterised in that：In step (1), if A fixed information source terminal S_kThe signal for needing to send is s_k(2t-1), k=1,2,3...K, then the transmission signal of K information source terminal It is with vector representation：S (2t-1)=[s₁(2t-1),...s_K(2t-1)]^T。

A kind of 3. virtual full duplex relaying transmission method according to claim 1 or claim 2, it is characterised in that： In step (1), R is relayed₂Carrying out the signal of K information source terminal that signal reception and detecting obtains in previous time slot (2t-2) can be with It is expressed as：After decoding and recompiling, then out-trunk R can be obtained₂ Being sent to the signal of K destination end can be expressed as：Wherein α₂Represent relaying R₂'s Power control ratio, i.e.,Mark computing is asked in wherein tr () expressions,Represent relaying R₂To K The channel coefficient matrix of destination end, its size are K × M, subscript ()^HRepresent conjugate transposition.

A kind of 4. virtual full duplex relaying transmission method according to claim 3, it is characterised in that：In step (1), in After R₁When receiving the signal from K information source terminal, relaying R now₁It can receive from K information source terminal and relaying R₂ The superposed signal of formation, then the superposed signal can be expressed as：

Wherein H_SR1Represent whole from K information source Hold relaying R₁Channel coefficient matrix, its size is M × K, G_R2R1Represent relaying R₂To relaying R₁Channel coefficient matrix its is big Small is M × M, z_R1(2t-1) represents relaying R₁White Gaussian noise.

A kind of 5. virtual full duplex relaying transmission method according to claim 4, it is characterised in that：In step (1), the K destination end D_kReceive from relaying R₂Signal be：WhereinRepresent from relaying R₂To k-th of terminal D_kChannel vector, its size is 1 × M, z_D,k(2t-1) represents k information source terminal D_kWhite Gaussian noise, k=1,2,3...K.

A kind of 6. virtual full duplex relaying transmission method according to claim 1, it is characterised in that：In step (2), if Determine information source terminal S_kThe signal for needing to send is s_k(2t), k=1,2,3...K, then the transmission signal of K information source terminal can use Vector is expressed as：S (2t)=[s₁(2t),....s_K(2t)]^T。

A kind of 7. virtual full duplex relaying transmission method according to claim 6, it is characterised in that：In step (2), in After R₁It can be expressed as in the signal that previous time slot (2t-1) carries out signal reception and detects K obtained information source terminal：After decoding and recompiling, then out-trunk R can be obtained₁It is sent to K The signal of destination end can be expressed as：α₁Represent relaying R₁Power control ratio, can To be expressed asMark computing is asked in wherein tr () expressions,Represent relaying R₁To K destination end Channel coefficient matrix, its size is K × M, subscript ()^HRepresent conjugate transposition.

A kind of 8. virtual full duplex relaying transmission method according to claim 6 or claim 7, it is characterised in that： In step (2), R is relayed₂When receiving the signal from K information source terminal, relaying R now₂It can receive from K information source Terminal and relaying R₁The superposed signal of formation, then the superposed signal can be expressed as：

Wherein H_SR2Represent from K information source terminal to relaying R₂ Channel coefficient matrix, its size is M × K, G_R1R2Represent relaying R₁To relaying R₂Channel coefficient matrix, its size is M × M, z_R2(2t) represents relaying R₂White Gaussian noise.

A kind of 9. virtual full duplex relaying transmission method according to claim 8, it is characterised in that：In step (2), the K destination end D_kReceive from relaying R₁Signal can be expressed as：WhereinRepresent from relaying R₁To k-th of information source terminal D_kChannel vector, its size is 1 × M, z_D,k(2t) represents that k information source is whole Hold D_kWhite Gaussian noise.