CN116318532A - RIS-SM-based signal detection method - Google Patents

Abstract

The invention relates to a signal detection method based on RIS-SM, which comprises the steps that a client side sends pilot frequency information to a base station side and selects a target transmitting antenna according to index bits; selecting a target modulation symbol from the modulation symbol set according to the modulation bit; transmitting the modulation symbol to the RIS through the UE-RIS channel through a target transmitting antenna, reflecting the modulation symbol, and transmitting the modulation symbol to the base station through the RIS-base station channel; the base station calculates a cascade estimation channel by using an LS algorithm according to the received signal and the pilot signal; calculating an optimal transmitting antenna index set by utilizing an improved maximum ratio combining algorithm MMRC according to the received signal and the cascade estimation channel; the base station combines the index of the transmitting antenna in the optimal transmitting antenna index set with the modulation symbols in all the modulation symbol sets to construct a tree decoder; and carrying out joint detection on the index of the target transmitting antenna and the modulation symbol by using a tree decoder and a maximum and minimum Euclidean distance criterion to restore information bits.

Description

RIS-SM-based signal detection method

Technical Field

The invention belongs to the technical field of communication, and particularly relates to a signal detection method based on RIS-SM.

Background

6G wireless communication is intended to meet the high rate and support large heterogeneous traffic demands of various applications. These requirements are important to support such things as augmented reality, smart healthcare, automatic vehicle control, dense cellular connectivity, etc. With the integration of multiple wireless access technologies in the future, massive intelligent equipment is accessed, and a new 6G-oriented heterogeneous network faces serious tests in terms of energy consumption, system capacity and reliability. In order to reduce power consumption and improve quality of service, communication through RIS has been proposed as a candidate technology for 6G. Meanwhile, spatial modulation (Spatial Modulation, SM) has a good balance between spectrum efficiency and energy efficiency, and can avoid the problems of inter-channel interference and inter-antenna synchronization existing in the MIMO system, and is widely studied in intelligent transportation systems (Intelligent Traffic System, ITS), internet of things and the like. The concept of RIS auxiliary communication is introduced into SM to be combined, so that the advantages of SM can be better exerted, and the defect of the MIMO system is overcome. The RIS-assisted spatial modulation (RIS assisted Spatial Modulation, RIS-SM) system is a promising candidate for future 6G heterogeneous network fusion techniques, and may be applied in systems with high reliability concerns like ITS. Therefore, research into RIS-SM systems is important.

The Maximum Likelihood (ML) algorithm is one of the classical signal detection algorithms. Although the ML algorithm has optimal detection performance, it needs to check all nodes, and the detection complexity is high, so that the algorithm is hard to be implemented in hardware. In addition, some known solutions that utilize ML criteria, such as QR decomposition for Sphere Decoding (SD) and M-order maximum likelihood decoding, maintain optimal performance, but still suffer from some drawbacks. For example, SD complexity is highly dependent on the initial value of the search radius: if the initial radius is chosen to be too small, the index and transmission symbols may not be properly detected. But if the search radius is chosen too large, the search space can become very large and the complexity remains high. Compared with SD, the QR decomposition technology of M-order maximum likelihood decoding has fixed throughput and is suitable for hardware realization. However, for higher order modulation schemes, there is still a high level of complexity.

Disclosure of Invention

In order to solve the problems that the existing signal detection method is high in complexity and the detection performance cannot be approximately optimal, the invention provides a signal detection method based on RIS-SM, which comprises the following steps:

s1: the client sends pilot frequency information to the base station

Splitting the information bit of each sending time slot into two parts of space bit and baseband bit; selecting a target transmitting antenna according to the index bit; selecting a target modulation symbol from the modulation symbol set according to the modulation bit;

s2: the client transmits the modulation symbol to the RIS through the UE-RIS channel by a target transmitting antenna, reflects the signal by the RIS and transmits the reflected signal to the base station through the RIS-base station channel; the base station acquires a receiving signal Y;

s3: the base station receives the signal Y and pilot signal

Computing a cascade estimation channel using LS algorithm>

And estimates the channel from the received signal Y and the cascade>

Calculating an optimal transmitting antenna index set by utilizing an improved maximum ratio combining algorithm MMRC;

s4: the base station combines the index of the transmitting antenna in the optimal transmitting antenna index set with the modulation symbols in all the modulation symbol sets to construct a tree decoder;

s5: the base station utilizes a tree decoder to jointly detect the index and the modulation symbol of a target transmitting antenna according to the received signal; and restoring the information bits according to the mapping relation between the information bits of each sending time slot and the index and modulation symbols of the target transmitting antenna.

The invention has at least the following beneficial effects

The invention reduces the number of nodes to be calculated by the tree search decoder with the minimum cumulative Euclidean distance. Meanwhile, L most likely transmitting antenna index sets are obtained according to MMRC, so that the search space is reduced, and compared with the traditional signal detection complexity is lower and the detection speed is higher.

Drawings

FIG. 1 is a flow chart of the method of the present invention;

FIG. 2 is a schematic diagram of a flow framework of the present invention;

FIG. 3 is a schematic diagram of a tree decoder according to the present invention;

FIG. 4 is a schematic diagram of a tree decoder according to an embodiment of the present invention;

FIG. 5 is a graph comparing the complexity of the signal detection at different transmit and receive antennas according to the present invention and conventional signal detection;

FIG. 6 illustrates BER analysis of the present invention and conventional signal detection in different configurations;

fig. 7 is a graph comparing the performance of the present invention and conventional signal detection.

Detailed Description

Other advantages and effects of the present invention will become apparent to those skilled in the art from the following disclosure, which describes the embodiments of the present invention with reference to specific examples. The invention may be practiced or carried out in other embodiments that depart from the specific details, and the details of the present description may be modified or varied from the spirit and scope of the present invention. It should be noted that the illustrations provided in the following embodiments merely illustrate the basic idea of the present invention by way of illustration, and the following embodiments and features in the embodiments may be combined with each other without conflict.

Referring to fig. 1, the present invention provides a signal detection method based on RIS-SM, which includes:

s1: the client sends pilot frequency information to the base station

Splitting the information bit of each sending time slot into two parts, namely index bit and modulation bit; selecting a target transmitting antenna according to the index bit; root of Chinese characterSelecting a target modulation symbol from the modulation symbol set according to the modulation bit;

information bits (log) in each transmission slot ₂ (N _t )+log ₂ (M)) split into index bits log ₂ (N _t ) And modulation bits log ₂ (M)，N _t Representing the number of transmit antennas and M represents the modulation order. For example, when the number of transmitting antennas N _t For 4, the modulation order M is 4, each group has 4 information bits, and assuming an information sequence 1001, this indicates that the selected transmit antenna index is 3, and the symbol is the 2 nd symbol of the M modulation symbols of normalized energy.

Referring to fig. 2, the client includes N _t With the transmitting antenna, the modulation symbol set has M modulation symbols, and the RIS is formed by N=N _hor N _ver A reflecting unit is formed of N _hor For the number of RIS reflecting elements per row, N _ver For each column of the number of RIS transmitting units, the RIS is controlled by a microcontroller.

S2: the client transmits the modulation symbol to the RIS through the UE-RIS channel by a target transmitting antenna, reflects the signal by the RIS and transmits the reflected signal to the base station through the RIS-base station channel; the base station acquires a receiving signal Y;

Y＝HX _l,s +n＝(H) _l s+n

wherein Y represents a received signal acquired by a base station, l represents an index of a target transmitting antenna, s represents a target modulation symbol, and N _t Indicating the number of transmit antennas owned by the client, H indicating the actual channel of the cascade between the UE-RIS-base station, (H) _l Column l representing H, N represents a mean of 0 and a variance of N ₀ Additive white gaussian noise of (c).

S3: the base station receives the signal Y and pilot signal

Computing a cascade estimation channel using LS algorithm>

And estimates the channel from the received signal Y and the cascade>

Calculating an optimal transmitting antenna index set by utilizing an improved maximum ratio combining algorithm MMRC;

representing pilot signal +.>

Representing a cascade estimated channel>

Representation->

Y represents the received signal.

Said estimating a channel from the received signal Y and the cascade

The calculating of the optimal transmit antenna index set using the modified maximum ratio combining algorithm MMRC includes:

s31: first estimating a channel from a received signal Y and a cascade

Calculating characteristic factors of the transmitting antenna;

wherein, xi _i Representing the characteristic factor of the ith transmit antenna,

representation levelThe co-estimated channel, Y, represents the received signal,

representing a cascade estimation channel matrix->

Is the i-th column element of (c).

S32: and selecting the transmitting antenna with the maximum L characteristic factors from the characteristic factors of the transmitting antennas as an optimal transmitting antenna set.

Wherein, xi _i I represents the modulus of the characteristic factor of the transmit antenna, χ represents the optimal transmit antenna set.

Referring to fig. 3, S4: the base station combines the index of the transmitting antenna in the optimal transmitting antenna index set with the modulation symbols in all the modulation symbol sets to construct a tree decoder;

dividing a tree decoder into N _r Each stage is provided with L multiplied by M nodes, and each branch corresponds to one detection combination;

s5: the base station utilizes a tree decoder to jointly detect the index and the modulation symbol of a target transmitting antenna according to the received signal; and restoring the information bits according to the mapping relation between the information bits of each transmission time slot and the index and modulation symbols of the target transmitting antenna.

S51: setting an initial access vector U of a tree decoder; calculating Euclidean distances of all nodes of a first stage in the tree decoder; the Euclidean distance of each node of the first stage is y ₁ And (3) with

A Euclidean distance between them;

i＝1,2,....,L

j＝1,2,....,M

wherein y is ₁ Representing the 1 st element in the received signal vector Y,

representing a cascade estimation channel->

Line 1, first _i Column element, access vector is u= [ U ] ₁ ,u ₂ ,...,u _i ,...,u _L×M ]A level tree representing that each node is currently located in a tree search;

since the algorithm needs to calculate the Euclidean distance of all nodes of the first stage of tree search, namely the accumulated Euclidean distance of the first 1 receiving antenna, the algorithm is initialized

The Euclidean distance vector of each node of the first level is expressed as

The euclidean distance of all nodes of the first stage in the tree decoder, that is, the euclidean distance of all possible combinations of L transmitting antennas and M symbols on the first receiving antenna.

S52: euclidean distance vector D of all nodes from first stage of tree decoder ₁ Finding an index position w where a node with the minimum Euclidean distance is located, updating the corresponding position of an access vector U of the tree decoder according to the index position w, and executing S53;

the updating the elements in the access vector U of the tree decoder comprises the following steps:

u _w ＝u _w +1,u _w ∈U

wherein, the access vector of the U tree decoder, U _w Representing the corresponding element at the index position w of the access vector U in the tree decoder.

The index position w where the node with the minimum Euclidean distance is located includes:

w＝arg min{D ₁ }

w＝1,2,...,L×M

wherein, w is shown in tableIndication D ₁ The index position of the smallest element in the L×M elements.

S53: judging whether a certain element in the updated access vector U is greater than N _r If so, S55 is performed, otherwise S54 is performed;

s54: euclidean distance vector D to each node of first level ₁ The node expansion of the next stage of the branch corresponding to the minimum element in (3) assumes that the antenna index and the transmission symbol corresponding to the minimum branch are l respectively _i Sum s _j At this time, the number of stages corresponding to the next-stage node is u _w Updating the index position w where the Euclidean distance minimum element is located

And performs S52;

wherein l _i Representing the index of the transmitting antenna corresponding to the branch, j represents the index of the transmitting symbol of the branch,

representing updated former u _w Cumulative Euclidean distance of root receive antenna, < >>

Representation matrix->

The (u) _w Line I _i Column elements.

S55: according to the access vector U being greater than N _r The index of the target transmit antenna and the target modulation symbol are found.

In an embodiment, when the index of the transmitting antenna is l _i The transmission symbol is s _j When the cumulative Euclidean distance of the first k antennas, namely the k-th stage of the branch is

Wherein, the liquid crystal display device comprises a liquid crystal display device,k，(l _i j) represents the position of the antenna index l in the tree search _i And transmitting symbol s _j Node of kth level under branch, < +.>

Representing the cumulative minimum Euclidean distance, y, corresponding to the node _i Representing the i-th element in the received signal vector Y. Therefore, the combination corresponding to the minimum value in the numerical value of the end node of each branch is the result of signal detection. In tree detection, there are L main branches, representing the L most likely transmit antennas selected by the MMRC, each main branch in turn has M branches, representing M possible modulation symbols, so there are l×m branches in total in the tree decoder, representing all combinations of L transmit antennas and M transmission symbols. M=2, n is listed in fig. 4 _t ＝4，N _r Tree search decoder for spatial modulation signal detection when=4, l=2. The process of building a tree search decoder is as follows: firstly, according to the MMRC algorithm in S31, 4 characteristic factors are obtained, the largest two of the characteristic factors are taken, and the indexes of the transmitting antennas respectively corresponding to the largest two characteristic factors are respectively l ₁ 、l ₂ . There are a total of 4 possible combinations of antenna indexes and transmission symbols, respectively (l) ₁ ，s ₁ )、(l ₁ ，s ₂ )、(l ₂ ，s ₁ ) And (l) ₂ ，s ₂ ) Each corresponding to 4 branches of the tree in the following figure. The depth of the tree is N _r Starting from the first level, each node in the tree represents the cumulative minimum Euclidean distance of the branch and the number of levels at which it is located.

Taking the tree search decoder of fig. 4 as an example, the values of each node in the tree are as follows, and the search procedure of the proposed tree search decoder is described below.

First, according to S51

Calculating cumulative Euclidean distance of all nodes of the first stage, i.e

And U is initialized to u= [1,1]Then execute S52, D ₁ The first element in (1) is the smallest, so U is updated to U= [2,1]. After that, S53 is executed, and since the stop condition is not satisfied, the updated D is expanded to the next node according to S54 ₁ The first element of (2) is smallest, at this time D ₁ ＝[0.4，0.2，0.5，1.6]. Continuing to execute the algorithm steps, U and D thereafter ₁ The method comprises the following steps of: u= [2,2,1,1 ]]，D ₁ ＝[0.4，0.3，0.5，1.6]；U＝[2，3，1，1]，D ₁ ＝[0.4，0.45，0.5，1.6]；U＝[3，3，1，1]，D ₁ ＝[0.8，0.45，0.5，1.6]；U＝[3，4，1，1]，D ₁ ＝[0.8，1.4，0.5，1.6]At this point, the stop condition has not been reached yet, and the algorithm continues to execute. At this time D ₁ The third element in (a) is the smallest, so U= [3,4,2,1 ]]，D ₁ ＝[0.8，1.4，1.5，1.6]Continuing to execute S51, U: [4,4,2,1]，D ₁ ＝[1.2，1.4，1.5，1.6]Continuing to execute S52, at this time, D ₁ The first element in (a) is the smallest, so U is updated to U= [5,4,2,1 ]]. Then executing S53 to judge that the first element in U is larger than N _r If the minimum accumulated Euclidean distance has been obtained for the branch where the element exists in the existing tree search graph, S55 is executed, the element corresponds to the corresponding antenna index l ₁ And transmitting symbol s ₁ The method is the method. Then according to the antenna index l ₁ And transmitting symbol s ₁ The bit sequence can be restored. From the above searching steps, it can be seen that in the above example, the maximum likelihood detection algorithm needs to calculate N _t ×M×N _r Compared with 32 nodes, the MMRC-based tree search decoder only needs to calculate the cumulative euclidean distance of 11 nodes, so that the calculation complexity is reduced. The tree search decoder is more effective for complexity reduction as the transmit antenna, receive antenna, and modulation order are larger.

Referring to FIG. 5, in this section, at least 10 is shown ⁶ The Monte Carlo simulation results of the Rayleigh fading channels are analyzed for the proposed detection algorithm in terms of complexity and bit error rate. In complexity analysis, the calculation nodes of tree search are taken as absolute complexity measurement indexes. In SM detection algorithm, N _t 、N _r M is the detectionIs an indispensable index. For ML, the detection will traverse all possible combinations, with an absolute complexity of C _ML ＝N _t N _r M。

FIGS. 5 (a) and 5 (b) analyze the computational complexity C of the relative ML decoding of the proposed algorithm and the RSD, greedy algorithm based on a tree search algorithm _rel The reduced complexity of the algorithms relative to ML decoding can be seen visually. The relative complexity is defined as the ratio of the absolute complexity of the two, i.e. the complexity of the proposed algorithm and the relative ML decoding of the RSD decoding is defined as

C∈{C _pro ,C _RSD ,C _Greedy }。

In simulation of fig. 5 (a), it can be seen that the detection complexity of the proposed algorithm is much smaller than RSD, ML, regardless of the number of antennas, modulation scheme. In fig. 5 (a), the modulation schemes are all 64QAM, and when the SNR is 4dB and the transmitting antenna and the receiving antenna are 4 and 16, respectively, the proposed algorithm can reduce the complexity by nearly 87% compared to ML, by nearly 12% compared to RSD, and by 7% compared to greedy algorithm. In fig. 5 (b), the proposed detection algorithm is still less complex than the ML, RSD algorithm detection, and more complex than the greedy algorithm detection based on tree search.

In the simulation of the detection algorithm, simulation results under different numbers of transmitting antennas, receiving antennas and modulation mode configurations are selected for the RIS-SM system. BER simulation results are shown in fig. 6.

In fig. 6, the performance impact of the transmit antenna, receive antenna, modulation scheme on the proposed algorithm is shown. It can be seen that the modulation scheme is the same as N _t In comparison with N _r The number of (c) has a greater impact on the system BER. This is because of N _r Corresponding to the depth of the search tree, when the depth of the tree is larger, the influence of noise on the detection result of the single receiving antenna is smaller, and the detection error probability is smaller. At snr=8, N _t In the case of=4, m=64, n=64, N _r Ber=8 can be reduced to N _r Close to 55% of =4. N (N) _t As the BER increases, so does the BER. At the same time, keep the number of days of transmitting and receivingThe BER increases as the modulation order increases. This is because, when the modulation order is increased, interference is more likely to occur between signal constellation points, thereby increasing the detection error probability.

In fig. 7, the performance of ML, RSD, greedy algorithm and proposed algorithm under different configurations is compared for different signal-to-noise ratios, where the RIS number n=64. It can be seen that the proposed algorithm and RSD decoder provide near optimal detection performance, consistent with expectations. The detection performance is close to the optimal ML detection, but the complexity is far smaller than that of the ML detection, which represents the advantages of the detection algorithm.

Finally, it is noted that the above-mentioned preferred embodiments are only intended to illustrate rather than limit the invention, and that, although the invention has been described in detail by means of the above-mentioned preferred embodiments, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the scope of the invention as defined by the appended claims.

Claims (6)

1. A RIS-SM based signal detection method, comprising:

s1: the client sends pilot frequency information to the base station

Splitting the information bit of each sending time slot into two parts, namely index bit and modulation bit; selecting a target transmitting antenna according to the index bit; selecting a target modulation symbol from the modulation symbol set according to the modulation bit;

s2: the client transmits the modulation symbol to the RIS through the UE-RIS channel by a target transmitting antenna, reflects the signal by the RIS and transmits the reflected signal to the base station through the RIS-base station channel; the base station acquires a receiving signal Y;

s3: the base station receives the signal Y and pilot signal

Computing a cascade estimation channel using LS algorithm>

And estimates the channel from the received signal Y and the cascade>

Calculating an optimal transmitting antenna index set by utilizing an improved maximum ratio combining algorithm MMRC;

s4: the base station combines the index of the transmitting antenna in the optimal transmitting antenna index set with the modulation symbols in all the modulation symbol sets to construct a tree decoder;

s5: the base station utilizes a tree decoder to jointly detect the index and the modulation symbol of a target transmitting antenna according to the received signal; and restoring the information bits according to the mapping relation between the information bits of each sending time slot and the index and modulation symbols of the target transmitting antenna.

2. The RIS-SM based signal detection method according to claim 1, wherein the receiving signal Y comprises:

Y＝HX _l，s +n＝(H) _l s+n

wherein Y represents a received signal acquired by a base station, l represents an index of a target transmitting antenna, s represents a target modulation symbol, and N _t Indicating the number of transmit antennas owned by the client, H indicating the actual channel of the cascade between the UE-RIS-base station, (H) _l Column l representing H, N represents a mean of 0 and a variance of N ₀ Additive white gaussian noise of (c).

3. The RIS-SM based signal detection method of claim 1, wherein the cascade estimation channel

Comprising the following steps:

representing pilot signal +.>

Representing a cascade estimated channel>

Representation->

Y represents the received signal.

4. The RIS-SM based signal detection method of claim 1, wherein the estimating the channel from the received signal Y and the cascade

The calculating of the optimal transmit antenna index set using the modified maximum ratio combining algorithm MMRC includes:

s31: first estimating a channel from a received signal Y and a cascade

Calculating characteristic factors of the transmitting antenna;

s32: and selecting the transmitting antenna with the maximum L characteristic factors from the characteristic factors of the transmitting antennas as an optimal transmitting antenna set.

5. The RIS-SM based signal detection method of claim 4, wherein the characteristic factors of the transmit antennas include:

wherein, xi _i Representing the characteristic factor of the ith transmit antenna,

representing the cascade estimated channel, Y representing the received signal,

representing a cascade estimation channel matrix->

Is the i-th column element of (c).

6. The RIS-SM based signal detection method of claim 1, wherein the joint detection of the index and modulation symbols of the target transmit antenna with a tree decoder based on the received signal comprises:

s51: setting an initial access vector U of a tree decoder; calculating Euclidean distances of all nodes of a first stage in the tree decoder; the Euclidean distance of each node of the first stage is y ₁ And (3) with

A Euclidean distance between them;

i＝1，2，....，L

j＝1，2，....，M

wherein y is ₁ Representing the 1 st element in the received signal vector Y,

representing a cascade estimation channel->

Line 1, first _i Column element, access vector is u= [ U ] ₁ ，u ₂ ，...，u _i ，...，u _L×M ]A level tree representing that each node is currently located in a tree search; the Euclidean distance vector of each node of the first level is expressed as +.>

S52: euclidean distance vector D of all nodes from first stage of tree decoder ₁ Finding an index position w where a node with the minimum Euclidean distance is located, updating the corresponding position of an access vector U of the tree decoder according to the index position w, and executing S53;

the updating the elements in the access vector U of the tree decoder comprises the following steps:

u _w ＝u _w +1，u _w ∈U

wherein, the access vector of the U tree decoder, U _w Representing the corresponding element at the index position w of the access vector U in the tree decoder.

The index position w where the node with the minimum Euclidean distance is located includes:

wherein w represents D ₁ The index position of the smallest element in the L×M elements.

S53: judging whether a certain element in the updated access vector U is greater than N _r If so, S55 is performed, otherwise S54 is performed;

s54: euclidean distance vector D to each node of first level ₁ The node expansion of the next stage of the branch corresponding to the minimum element in (3) assumes that the antenna index and the transmission symbol corresponding to the minimum branch are l respectively _i Sum s _j At this time, the number of stages corresponding to the next-stage node is u _w Updating the index position w where the Euclidean distance minimum element is located

And performs step S52;

wherein l _i Representing the index of the transmitting antenna corresponding to the branch, j represents the index of the transmitting symbol of the branch,

representing updated former u _w Cumulative Euclidean distance of root receive antenna, < >>

Representation matrix->

The (u) _w Line I _i Elements of a column;

s55: according to the access vector U being greater than N _r The index of the target transmit antenna and the mapped symbol are found.

Images