阅读说明：本技术 一种无线光通信中的完全广义空间调制方法 (Complete generalized spatial modulation method in wireless optical communication ) 是由 王惠琴 杨顺信 张悦 王燕 曹明华 毛一聪 叶归清 侯文斌 于 2020-04-10 设计创作，主要内容包括：一种无线光通信中的完全广义空间调制方法,用于数据传输的激光器从仅激活一个激光器到激活多个甚至所有激光器而变化,这使得传输速率与激光器数量呈线性比例关系,显著提高了系统的传输速率。同时,为了充分利用激光器组合冗余,采用基于信道范数最大化选择算法完成了激活激光器的选择,有效改善了系统的误码性能。在此基础上利用联合界技术推导了该方法在对数正态信道下的理论误码率上界。相对于传统的光空间调制,本发明不仅实现了传输速率和激光器数量之间的线性关系,且解决了激光器数必须为2的整数次幂的问题,显著提升了系统的传输速率。同时,通过激光器选择算法的引入,有效的改善了系统的误码性能。(A method for completely generalized spatial modulation in wireless optical communication is provided, wherein a laser used for data transmission is changed from the activation of only one laser to the activation of a plurality of lasers or even all lasers, so that the transmission rate is in a linear proportional relation with the number of the lasers, and the transmission rate of a system is obviously improved. Meanwhile, in order to fully utilize the laser combination redundancy, the selection of the activated laser is completed by adopting a channel norm maximization selection algorithm, and the error code performance of the system is effectively improved. On the basis, the theoretical bit error rate upper bound of the method under the lognormal channel is deduced by using a joint bound technology. Compared with the traditional optical spatial modulation, the invention not only realizes the linear relation between the transmission rate and the number of the lasers, but also solves the problem that the number of the lasers must be an integer power of 2, and obviously improves the transmission rate of the system. Meanwhile, the error code performance of the system is effectively improved by introducing a laser selection algorithm.)

1. A method of fully generalized spatial modulation in wireless optical communications, characterized in that a linear relationship between transmission rate and number of lasers is achieved by activating different numbers of lasers to communicate information. Meanwhile, in order to fully utilize the laser combination redundancy, an antenna selection algorithm is introduced, and the selection of the activated laser is completed by adopting a channel norm maximization-based selection algorithm, so that the error code performance of the system is effectively improved. On the basis, the theoretical bit error rate upper bound of the method under the lognormal channel is deduced by using a joint bound technology.

2. The method according to claim 1, comprising the following steps:

step 1: suppose there is N in the system_tA laser, N_rA detector, and the emitting end activates N each time_u(N_u＝1，……,N_t) A laser, so from N_tActivating N in one laser_uThe number of alternative combinations ofAnd (4) seed preparation. Since the number of laser combinations used must be an integer power of 2, it is only necessary to operate fromSelected from a variety of alternative combinationsAnd (5) planting. Since a subset of lasers ranges from activating only one laser to activating a plurality or all of the lasersVaries so that the laser index carries bits of information ofAssuming an L-order PAM modulation is used, each PAM modulation symbol carries log₂(L) bit information. Thus, it has a transmission rate of

The transmitting terminal utilizes the characteristics of the laser combination mapping vector and the L-PAM modulation symbol to construct a transmitting signal matrix:in the formula (I), the compound is shown in the specification,is a number N_t× 1 dimensional laser mapping vector, where e is 1 ≦ e < f ≦ N_tAnd the position of the non-zero element represents the sequence number of the active laser.(1. ltoreq. L. ltoreq.L denotes the amplitude of the transmitted pulse) is the mapping vector of L-PAM.

Step 2: assuming that the current channel state information is known at both the transmitting and receiving ends, the transmitting end can select a suitable laser combination according to the current channel state information and a laser selection algorithm, and complete the mapping of corresponding information. Wherein the laser selection algorithm employs a method based on maximizing the channel norm to select the active laser. The basic idea is as follows: the subchannel with the largest channel norm is selected to transmit information, and the essence is that the received signal-to-noise ratio is largest. The corresponding selection algorithm is

Where P denotes the set of all possible laser subsets and P denotes the set of laser subsets selected according to the algorithm.

And step 3: setting the output signal of the light detector as

In the formula (I), the compound is shown in the specification,is a mean of zero and a variance ofη is the photoelectric conversion efficiency, H is N_r×N_tThe channel fading matrix is dimensional, and the channel fading coefficient h is distributed according to the log normal distribution.

Assuming that the receiving end knows the channel state information, i.e. under the condition that H is known, we use the maximum likelihood decoding algorithm to perform signal detection, and recover the original bit information.

Wherein | · | purple sweet_FThe expression is given in the F-norm,andrespectively representing the sequence number at which the active laser was detected and the modulation symbol transmitted.

And 4, step 4: in the F-GSM constellation space, the euclidean distance between constellation points of the transmission signal is a decisive factor for the probability of system errors. The theoretical upper bound of the F-GSM error rate is obtained by the joint bound technique. The specific method comprises the following steps:

suppose the transmission rate of F-GSM is R_F-GSMbpcu, from each transmitting endA possible transmission signal x₁,x₂,L,One signal is selected for transmission. The upper bound of the F-GSM system bit error rate theory when the ML detection algorithm is adopted can be obtained through the united bound technology

Wherein d is_H(x_i,x_j) Denotes x_iAnd x_jHamming distance between, i.e. signal x_iThe error decision is x_jThe number of error bits generated. P (x)_i→x_jH) represents that the receiving end knows the channel matrix H, the symbol x is sent_iIs erroneously detected as a symbol x_jPair-wise error probability.

Technical Field

The invention relates to a completely generalized spatial modulation method in wireless optical communication, which transmits information by activating different numbers of lasers, thereby realizing the linear relation between the transmission rate and the number of the lasers, and belonging to the technical field of wireless optical communication.

Background

The atmospheric laser communication has attracted extensive attention of researchers due to the advantages of high transmission rate, good confidentiality, no need of frequency license and the like. Although the frequency of the optical carrier is high, the bandwidth of the optical signal can exceed 1 THz. In actual use, however, transmission rates are often limited to 10Gb/s or less due to limitations in atmospheric attenuation, pointing errors, atmospheric turbulence, and electronic device speeds. It is difficult to satisfy the demands for high-speed, large-capacity communication brought by the full-service and rapid development of the internet. How to improve the transmission rate of the atmospheric laser communication system and accelerate the popularization and application of the atmospheric laser communication system is urgent.

In recent years, Optical Spatial Modulation (OSM) has been developed to improve the transmission rate of atmospheric laser communication and to resist the effect of atmospheric turbulence. The optical multi-Input multi-Output (OMIMO) technology is a novel optical multi-Input multi-Output (OMIMO) technology, and can effectively solve the defects in the traditional OMIMO. Optical spatial modulation not only uses a conventional digital modulation constellation (i.e., signal domain) to convey information, but also carries information in addition through the laser index (i.e., spatial domain). In this way, the laser not only forms the medium of the wireless link, but also carries the information itself. Therefore, it provides an effective measure for increasing the transmission rate of the system. In view of this, the researchers have conducted extensive research on how to design efficient spatial modulation of light.

Early optical spatial modulation originated from indoor visible light communication. Thilo Fath et al propose different optical spatial modulations using Pulse Amplitude Modulation (PAM), and achieve an increase in transmission rate by improving spectral efficiency. However, high-order PAM modulation is very sensitive to receiver noise and atmospheric turbulence, so that the error code performance is not ideal. In view of this, Wasiu o.popola et al construct optical spatial modulation using Pulse Position Modulation (PPM) and obtain better error performance. However, when the modulation order is high, the difficulty of ensuring strict synchronization between the laser and the detector is greatly increased. In the research, only one laser is activated during each transmission, which effectively solves the problems of inter-channel interference, difficult laser synchronization and the like, but obviously limits the utilization rate of space resources because only one laser is activated each time; it also requires that the number of lasers must be an integer power of 2. To this end, s.p. alaka and Hammed g.olanrewaju have constructed different Optical Generalized Spatial Modulation (OGSM) by simultaneously activating a small number of lasers using PAM and PPM Modulation, respectively. The scheme improves the transmission rate of the system by improving the utilization rate of space resources, and simultaneously solves the problem that the number of lasers must be an integer power of 2.

Although the above studies are each characterized, they share a common deficiency in that the transmission rate is logarithmically proportional to the number of lasers. This makes the increase of the transmission rate of the system limited, and there is a large gap compared with the transmission rate of Vertical-layered space-time codes (V-BLAST). To obtain the same transmission rate as V-BLAST, the OSM system needs to add extra laser, resulting in waste of space resources. Based on this, the invention provides a method of Fully Generalized Spatial Modulation (F-GSM) in wireless optical communication by Fully utilizing all possible laser combinations under the condition that the total number of lasers is not changed. The method realizes the linear relation between the transmission rate and the number of lasers, and solves the problem that the number of lasers must be an integer power of 2. Meanwhile, the error code performance of the F-GSM system is effectively improved by selecting and activating the laser based on the algorithm of channel norm maximization.

Disclosure of Invention

The invention aims to establish a completely generalized spatial modulation method in wireless optical communication by activating different numbers of lasers, so as to realize a linear relation between a transmission rate and the number of the lasers, thereby improving the transmission rate of a wireless optical communication system.

The invention relates to a method for improving transmission rate in wireless optical communication. Characterized in that information is transferred by activating different numbers of lasers, whereby a linear relationship between the transmission rate and the number of lasers is achieved. Meanwhile, in order to fully utilize the laser combination redundancy, the selection of the activated laser is completed by adopting a channel norm maximization selection algorithm, and the error code performance of the system is effectively improved. On the basis, the theoretical bit error rate upper bound of the method under the lognormal channel is deduced by using a joint bound technology. The method comprises the following specific steps:

step 1: suppose there is N in the system_tA laser, N_rA detector, and the emitting end activates N each time_u(N_u＝1，……,N_t) A laser, so from N_tActivating N in one laser_uThe number of alternative combinations ofAnd (4) seed preparation. Since the number of laser combinations used must be an integer power of 2, it is only necessary to operate fromSelected from a variety of alternative combinationsAnd (5) planting. Since the subset of lasers varies from activating only one laser to activating multiple or all lasers, the information bits carried by the laser indexIs composed ofAssuming an L-order PAM modulation is used, each PAM modulation symbol carries log₂(L) bit information. Thus, it has a transmission rate of

The transmitting terminal utilizes the characteristics of the laser combination mapping vector and the L-PAM modulation symbol to construct a transmitting signal matrix:in the formula (I), the compound is shown in the specification,is a number N_t× 1 dimensional laser mapping vector, where e is 1 ≦ e < f ≦ N_tAnd the position of the non-zero element represents the sequence number of the active laser.(1. ltoreq. L. ltoreq.L denotes the amplitude of the transmitted pulse) is the mapping vector of L-PAM.

Step 2: assuming that the current channel state information is known at both the transmitting and receiving ends, the transmitting end can select a suitable laser combination according to the current channel state information and a laser selection algorithm, and complete the mapping of corresponding information. Wherein the laser selection algorithm employs a method based on maximizing the channel norm to select the active laser. The basic idea is as follows: the subchannel with the largest channel norm is selected to transmit information, and the essence is that the received signal-to-noise ratio is largest. The corresponding selection algorithm is

Where P denotes the set of all possible laser subsets and P denotes the set of laser subsets selected according to the algorithm.

And step 3: setting the output signal of the light detector as

In the formula (I), the compound is shown in the specification,is a mean of zero and a variance ofη is the photoelectric conversion efficiency, H is N_r×N_tThe channel fading matrix is dimensional, and the channel fading coefficient h is distributed according to the log normal distribution.

Assuming that the receiving end knows the channel state information, i.e. under the condition that H is known, we use the maximum likelihood decoding algorithm to perform signal detection, and recover the original bit information.

Wherein | · | purple sweet_FThe expression is given in the F-norm,andrespectively representing the sequence number at which the active laser was detected and the modulation symbol transmitted. And after the signal detection is finished, demapping according to the mapping rule.

And 4, step 4: in the F-GSM constellation space, the euclidean distance between constellation points of the transmission signal is a decisive factor for the probability of system errors. The theoretical upper bound of the F-GSM error rate is obtained by the joint bound technique. The specific method comprises the following steps:

assume that the transmission rate of F-SM is R_F-GSMbits/s, per transmission from the masterA possibility ofOf the transmitting signalOne signal is selected for transmission. The upper bound of the F-GSM system bit error rate theory when the ML detection algorithm is adopted can be obtained through the united bound technology

Wherein d is_H(x_i,x_j) Denotes x_iAnd x_jHamming distance between, i.e. signal x_iThe error decision is x_jThe number of error bits generated. P (x)_i→x_jH) represents that the receiving end knows the channel matrix H, the symbol x is sent_iIs erroneously detected as a symbol x_jPair-wise error probability.

The invention has the advantages that: a completely generalized spatial modulation method in wireless optical communication is provided by activating different numbers of lasers, so that a linear relation between a transmission rate and the number of lasers is realized, and the problem that the number of lasers must be an integer power of 2 is solved. Meanwhile, the combination redundancy of the lasers is fully utilized, the selection of the activated lasers is completed by adopting a channel norm maximization-based selection algorithm, and the error code performance of the system is effectively improved. This provides an effective measure for constructing a large-capacity, high-rate optical spatial modulation system. Has certain reference value.

Drawings

FIG. 1 shows a signal transmission model of F-GSM system in wireless optical communication

FIG. 2 is a flow chart of the method of the present invention

FIG. 3 shows the theoretical error rate and simulated error rate of the F-GSM system before the laser selection algorithm is adopted

FIG. 4 is a diagram showing the bit error rate of different F-GSM systems with the same transmission rate

FIG. 5 shows the bit error rates of different OSM schemes with the same transmission rate

For a more complete description of the embodiments, reference is now made to the accompanying drawings in which:

Detailed Description

The invention provides a completely generalized spatial modulation method in wireless optical communication, which transmits information by activating different numbers of lasers, thereby realizing the linear relation between the transmission rate and the number of lasers. Meanwhile, in order to fully utilize the laser combination redundancy, an antenna selection algorithm is introduced, and the selection of the activated laser is completed by adopting a channel norm maximization-based selection algorithm, so that the error code performance of the system is effectively improved. The present invention will be described in detail below with reference to specific embodiments thereof.

The invention is achieved by the following technical measures:

the invention relates to a completely generalized spatial modulation method in wireless optical communication, aiming at constructing a high transmission rate method suitable for the wireless optical communication by activating different numbers of lasers. Meanwhile, the selection of the active laser is completed by adopting a channel norm maximization-based selection algorithm, the error code performance of the system is effectively improved, and the upper bound of the theoretical error code rate of the method is deduced.

Step 1: the F-GSM system is shown in fig. 1. Suppose there is N in the system_tA laser, N_rA detector, and the emitting end activates N each time_u(N_u＝1，……,N_t) A laser, so from N_tActivating N in one laser_uThe number of alternative combinations ofAnd (4) seed preparation. Since the number of laser combinations used must be an integer power of 2, it is only necessary to operate fromSelected from a variety of alternative combinationsAnd (5) planting. When the active laser is selected, PAM modulation symbols can be loaded on the active laser at the same time for transmission.

In this system, sinceThe subset of lasers varies from activating only one laser to activating more or all of the lasers, so the laser index carries bits of information asAssuming an L-order PAM modulation is used, each PAM modulation symbol carries log₂(L) bit information. Thus, it has a transmission rate of

As can be seen from equation (1), the transmission rate of the F-GSM system is linearly proportional to the number of lasers, which significantly increases the transmission rate of the system. But when we look carefully at the alternative set of this scheme we find that it has some redundancy, i.e. it is redundantSeed redundancy combinations are not adopted; and when N is_tThe larger the value, the more redundant it is. Therefore, how to reasonably select the combination from the candidate set is a key. Most of the existing optical spatial modulation schemes adopt a random selection method to select laser combinations, which limits the error code performance of the system to some extent. Notably, the atmosphere is a time-varying fading channel. That is, the optical signal experiences different effects as it passes through different sub-channels. Therefore, the combination of the active lasers is selected according to the quality of the current channel state, namely, the active lasers are selected by adopting a channel norm maximization-based method, and the method can obviously reduce the bit error rate of the system.

Step 2: assuming that the current channel state information is known at both the transmitting and receiving ends, the transmitting end can select a suitable laser combination according to the current channel state information and a laser selection algorithm, and complete the mapping of corresponding information. Wherein the laser selection algorithm employs a method based on maximizing the channel norm to select the active laser. The basic idea is as follows: the subchannel with the largest channel norm is selected to transmit information, and the essence is that the received signal-to-noise ratio is largest. The corresponding selection algorithm is

Where P denotes the set of all possible laser subsets and P denotes the set of laser subsets selected according to the algorithm.

And step 3: at a transmitting end, mapping an F-GSM signal, specifically comprising the following steps:

if the input binary bit stream is b, the binary bit stream is divided into b after serial-to-parallel conversion₁And b₂Two parts. Wherein, b₁Is mapped as an index of the active lasers in the system, and the mapping relationship can be N_tVector x of dimension × 1_LaserTo indicate that isWherein e is more than or equal to 1 and f is more than or equal to N_tThe position of the non-zero element represents the index of the active laser, [ g [ ]]^TRepresenting a transpose operation. b₂Is mapped to a certain modulation symbol in the conventional L-PAM modulation constellation diagram, and the mapping relation can be expressed asWherein, I_oRepresenting the average light intensity of the transmitted L-PAM modulation symbols. Assuming that the average power distribution mechanism is adopted at the transmitting end, when the number of the active lasers is N_uWith total power of transmission per active laserAnd (4) doubling. The transmission signal of the F-GSM system can be expressed as

Wherein the total power transmitted satisfies E (xx)^H)＝1，[g]^HRepresenting a conjugate transpose operation. Suppose the transmission rate of the system is 3bpcu, where the number of lasers N_t3, modulation order L2, and current channel state matrix is

In this case, the alternative set of all laser combinations in the F-GSM system is P { (1), (2), (3), (1,2), (1,3), (2,3), (1,2,3) }, the channel norms of all possible combinations are calculated, and 2 with the largest norm is selected by using the selected antenna selection method^Nt-1Information is transferred by 4 laser combinations, i.e., p { (1,2,3), (1,3), (1,2), (1) }.

And 4, step 4: after the signal modulation of the transmitting end is completed, the modulated signal is received by the optical detector after passing through the atmospheric channel. Setting the output signal of the light detector as

In the formula (I), the compound is shown in the specification,is a mean of zero and a variance ofη is the photoelectric conversion efficiency, H is N_r×N_tThe channel fading matrix is maintained. For weak to moderate atmospheric turbulence, the channel fading coefficient h is usually distributed from lognormal with probability density function of

In the formula, if the fading strength is normalized, it is E [ h]1, then mean valueVariance (variance)Wherein the content of the first and second substances,a flicker index representing the variation of channel fading.

At the receiving end, it is a key to correctly detect the signal sent by the sending end from the output signal of the detector. Currently, a commonly used method is a Maximum Likelihood (ML) decoding algorithm. The ML decoding algorithm detects the index and modulation symbol of the active laser by an exhaustive search method with the criterion of

Wherein | · | purple sweet_FThe expression is given in the F-norm,andrespectively representing the sequence number at which the active laser was detected and the modulation symbol transmitted. Recovering the original bit after the output result of the formula (7) is subjected to F-GSM demapping

And 5: in order to verify the correctness of the invention, the upper bound of the theoretical bit error rate under the maximum likelihood detection is deduced, and the specific steps are as follows:

in the F-GSM constellation space, the euclidean distance between constellation points of the transmission signal is a decisive factor for the probability of system errors. The theoretical upper bound of the F-GSM bit error rate is obtained here by the joint bound technique.

Assuming that the receiving end knows the channel state matrix H, a symbol x is transmitted_iIs erroneously detected as a symbol x_jHas a pair-wise error probability of

By bringing formula (5) into (8)

Wherein the content of the first and second substances,subject to mean E (D) 0, varianceGaussian random variable of (2). Thus, equation (9) can be written as

Wherein the content of the first and second substances,it can be brought into the formula (10)

Therefore, the upper bound of the theoretical bit error rate of the F-GSM system is as follows when the ML detection algorithm is adopted can be obtained by the combined bound technology

Wherein the content of the first and second substances,representing all possible sets of transmitted symbols, d, in the F-GSM system_H(x_i,x_j) Denotes the transmitted symbol x_iAnd detecting the symbol x_jHamming distance between. Observing (12) formula, the bit error rate BER and channel state matrix H and transmission rate R of F-GSM system can be easily found_F-GSMAnd the like. At R_F-GSMUnder certain conditions, the larger the H, the smaller the BER of the F-GSM system.

To verify the correctness of the scheme, we performed simulation analysis using a monte carlo simulation method. The simulation conditions are as follows: assuming that the receiving end knows the channel state information, the total power of the system is unchanged, the power of each active laser is evenly distributed, the receiving end adopts ML detection algorithm and flicker indexThe photoelectric conversion efficiency η is 0.5, and GSPAM activates only two lasers.

FIG. 3 shows the theoretical error rate and simulated error rate of the F-GSM system before the laser selection algorithm is used. Wherein, the modulation mode adopts 4-PAM. As can be seen from the simulation results: when the signal-to-noise ratio is low, the actual bit error rate of the F-GSM system is lower than the theoretical upper bound; and when the signal-to-noise ratio is large, the two are basically superposed.

Fig. 4 shows the bit error rates of different F-GSM systems with the same transmission rate. Wherein the transmission rate is 5 bpcu. As can be seen from the simulation results: under the condition of the same transmission rate, increasing the number of lasers (namely, reducing the modulation order of PAM) can improve the error code performance of the front F-GSM system and the rear F-GSM system by adopting a laser selection algorithm. Compared with the F-GSM system before the laser selection algorithm is adopted, the F-GSM system error code performance is improved more obviously after the laser selection algorithm is adopted, and the more the number of lasers is, the more the performance is improved.

Fig. 5 shows the error rates of different OSM schemes with the same transmission rate. At this time, the transmission rate is 5 bpcu. As can be seen from the simulation results: firstly, when the number of lasers is the same and the modulation orders are different, under the condition of small signal to noise ratio, the error rate of the F-GSM scheme before the adoption of a laser selection algorithm is between SPAM and GSPAM; the error rate of the F-GSM scheme is slightly lower than that of SPAM and GSPAM under the condition of large signal-to-noise ratio. Secondly, when the modulation orders are the same and the number of lasers is different, under the condition of a small signal to noise ratio, the error rate of the F-GSM scheme is slightly higher than that of SPAM and GSPAM; under the condition of large signal-to-noise ratio, the error rate of the F-GSM scheme is smaller than that of the GSPAM scheme and is slightly larger than that of the SPAM scheme. And thirdly, the error code performance of the F-GSM scheme is obviously superior to that of other schemes after the laser selection algorithm is adopted.

From the above description of the embodiments, it is clear for a person skilled in the art that the present invention can be implemented in software or by hardware. Based on the above understanding, the contribution of the technical method of the present invention to the prior art can be partially performed by software or hardware.