A power control method for uplink non-orthogonal multiple access short packet communication
By coordinating user transmission power and transmission rate on the base station side, using extremely large and extremely small optimization problems and dual search algorithms, the user fairness problem in uplink non-orthogonal multi-access short packet communication is solved, and the system performance and service quality are improved. , suitable for IoT applications.
Patent Information
- Application Number
- CN202211509810.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-11-29
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2042-11-29
AI Technical Summary
The prior art lacks an uplink non-orthogonal multi-access short-packet communication power control method based on user fairness, resulting in uneven power control among users in Internet of Things applications, affecting system performance and service quality.
Through the base station coordinates the user's transmission power and transmission rate based on channel status information and user transmission parameters, and adopts extremely large and extremely small optimization problems and dual search algorithms to achieve uplink non-orthogonal multi-access short packet communication power control for users, maximizing and minimizing User effective throughput.
It improves the overall performance of the system, ensures service quality, meets the needs of massive machine-type communications and ultra-reliable low-latency communications in the Internet of Things, and provides a power control solution with the best user fairness.
Smart Images

Figure CN115802324B_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of wireless communications, and in particular relates to a power control method for uplink non-orthogonal multiple access short packet communications. Background Art
[0002] With the application and development of the fifth generation of mobile communication technology, the Internet of Things technology has been gradually applied to various industries to support massive machine type communications (mMTC). To adapt to the explosive data growth and access requirements, non-orthogonal multiple access (NOMA) technology is a feasible solution to increase the number of accesses and thus increase spectrum efficiency. NOMA is achieved by allowing multiple users to access the same resource blocks, such as time slots, frequency bands, etc. In power domain NOMA, successive interference cancellation (SIC) can be used to estimate the signal and is usually performed based on the decoding order of the channel state information (CSI).
[0003] In addition, the main applications of IoT are real-time interactive systems, such as industrial automation, intelligent transportation, and augmented / virtual reality, which require ultra-reliable and low-latency communications (URLLC). In order to reduce transmission delay, short packet communication (SPC) is needed, that is, communication is performed using a limited transmission block length. In short packet communication, there is a transmission error rate in wireless transmission. Existing studies have discussed the system model of short packet communication and given an approximate relationship between the achievable rate and the error rate and transmission block length.
[0004] Although there have been many studies on the throughput of downlink NOMA systems using a finite block length mechanism, it is equally important to analyze the power control methods for uplink NOMA short packet communications due to the important differences between uplink and downlink in terms of inter-user interference, SIC decoding order, and resource allocation strategies. More importantly, due to the complexity of short packet communications, most studies on user fairness-based power control in NOMA systems focus on the infinite block length mechanism. To simplify the problem, in general studies of two-user NOMA short packet communications, the effective throughput of one user is often used as the optimization target, while the effective throughput requirement of the other user is regarded as a quality of service (QoS) constraint. On this basis, how to perform system power control is explored. This simplification ignores the global regulation that satisfies user fairness, but only targets the power control scheme for a single user.
[0005] In summary, how to control the power of uplink non-orthogonal multiple access short packet communication is an important issue that needs to be solved in current IoT applications. The lack of power control methods based on user fairness has become an important factor restricting the development of IoT related technologies. Summary of the invention
[0006] The present invention aims to provide a power control method for uplink non-orthogonal multiple access short packet communication, which can realize power control of an uplink non-orthogonal multiple access short packet communication system based on user fairness.
[0007] In order to solve the above technical problems, the specific technical solutions of the present invention are as follows:
[0008] A power control method for uplink non-orthogonal multiple access short packet communication includes a base station coordinating the transmission schemes of each user according to the transmission reliability requirements and hardware limitations of the non-orthogonal multiple access users, and the uplink non-orthogonal multiple access short packet communication process is supervised by a power control method based on service quality and actual constraints.
[0009] The power control method steps are as follows:
[0010] Step 1: Before the transmission starts, the user connects to the base station for pairing, and transmits the non-orthogonal multiple access short packet transmission parameters to the base station via the uplink;
[0011] Step 2: The base station receives the non-orthogonal multiple access short packet transmission parameters, and estimates the transmission power and short packet transmission rate of each user based on the known channel state information between each user and the base station to maximize the minimum user effective throughput;
[0012] Step 3: The base station feeds back the estimated transmission power and short packet transmission rate to each user;
[0013] Step 4: Each user transmits a short packet to the base station at the optimal transmission power and transmission rate, completing the uplink non-orthogonal multiple access short packet communication process.
[0014] Furthermore, the non-orthogonal multiple access short packet transmission parameters in step 1 include the maximum allowed block error rate, the short packet transmission block length, and the upper limit of the maximum transmission power of the user equipment.
[0015] Furthermore, step 2 specifically includes the following steps:
[0016] Step 2.1, establish a minimax optimization problem according to the non-orthogonal multiple access short packet transmission parameters, introduce auxiliary variables to transform the objective function into constraints, and simplify the original non-convex optimization problem; initialize the block length, the power search accuracy of the outer loop, the transmission rate search accuracy of the inner loop, and the iteration termination condition of the inner loop;
[0017] Step 2.2, in the outer loop, set the initial value for the one-dimensional search of the transmit power;
[0018] Step 2.3, in the inner loop, using an iterative algorithm based on fixed interval search to solve the optimal value of the problem under the given transmission power;
[0019] Step 2.4: Determine the absolute value of the difference between the auxiliary variables of two adjacent iterations. If it is less than the initial setting value of the iteration termination condition, it is determined that the iteration converges to the optimal value and the maximized minimum user effective throughput is obtained, and then proceed to step 2.5; otherwise, return to step 2.3;
[0020] Step 2.5: In the outer loop, determine whether the optimal value obtained in step 2.4 is the global optimal value. If it is greater than the existing global optimal value, replace it; otherwise, discard it directly without replacement;
[0021] Step 2.6, update the transmission power set by the outer loop, and return to step 2.3 until the one-dimensional search process is completed for the power within the feasible range with the initialized power search accuracy;
[0022] Step 2.7: Obtain the optimal transmission power corresponding to the global optimal value, and calculate the optimal transmission rate.
[0023] Furthermore, the minimax optimization problem described in step 2.1 is established and auxiliary variables are introduced to transform the objective function into constraints, as follows:
[0024] Assume an uplink non-orthogonal multiple access system consisting of a base station and two users. Both the base station and the users have single antennas. The transmission signals of the two users are where s i is the transmitted signal, p i is the transmission power; the signal received by the base station is expressed as where h i is the channel gain, set |h1| 2 >|h2| 2 , w means the mean is zero and the variance is σ 2 Additive Gaussian white noise;
[0025] Successive interference cancellation is used to separate signals at the base station side; Unlike the decoding order that is always fixed in the downlink non-orthogonal multiple access system, the decoding order at the base station side is arbitrary for the uplink system because the base station will receive signals from all users. There are many factors that affect the decoding order of the base station, such as channel gain, available resource limitations, and QoS constraints. The present invention sets a decoding order based on channel state information, that is, signals with high channel gain will be decoded first at the base station;
[0026] Therefore, at the base station side, the signal of user 1 is first demodulated, and its signal-to-interference-to-noise ratio is
[0027] According to existing research, the transmission rate R, that is, the amount of effective data sent per unit bandwidth per second, is approximately:
[0028]
[0029] Where γ represents the signal-to-interference-noise ratio, n represents the block length, ε represents the block error rate, C(γ)=log(1+γ) is the Shannon capacity, V(γ)=(1-(1+γ) -2 )(loge) 2 is the channel dispersion, Q -1 (·) is the inverse transform of the Gaussian Q function;
[0030] In a non-orthogonal multiple access short packet communication system, there is a non-negligible error in demodulation, so the probability of correctly demodulating s1 at the base station is (1-ε1), where ε1 = Q(f(γ1,n,R1)), which can be obtained by transforming formula (10);
[0031] If s2 is set to be correctly demodulated only after s1 is correctly demodulated during the SIC process, then when the base station demodulates s2, its signal to interference noise ratio is The overall decoding error rate of s2 is expressed as where ε2=Q(f(γ2,n,R2));
[0032] The effective throughput index of uplink non-orthogonal multiple access short packet communication is:
[0033]
[0034] Regarding user fairness, there is a minimax problem, in which power and transmission rate are optimization variables, and the minimum user effective throughput is the objective function. Maximizing the objective function is the optimization direction of the problem, as shown below:
[0035]
[0036] Among them, C0 is the power constraint, C1 and C2 are transmission reliability constraints, and C3 is the practical feasibility constraint;
[0037] because and Therefore, no matter min{T1,T2}=T1 or T2, p1=P max In addition, the following formula is obtained:
[0038]
[0039] According to formula (13), since ε increases monotonically with R, ε max With R max The corresponding conditions are consistent; therefore, the constraints in problem (12) are transformed to obtain the following problem:
[0040]
[0041] in,
[0042] Solving problem (14) is divided into two stages. The first stage is to find the optimal value under the condition of fixed transmission power p2. The second stage is to perform a one-dimensional search for the transmission power p2, obtain the optimal solution of power p2 by comparing the optimal value obtained in the first stage, and calculate the optimal solution of transmission rates R1 and R2.
[0043] The issues that need to be addressed in the first phase are as follows:
[0044]
[0045] In order to solve problem (15), an auxiliary variable t is introduced to transform the objective function in the minimax problem into a constraint. The transformed problem is as follows:
[0046]
[0047] Furthermore, step 2.3 is as follows:
[0048] First, for each fixed value of t, verify whether the problem has a feasible solution, that is, it satisfies all constraints at the same time; there is a theorem: R i [1-Q(f(γ i ,n,R i ))], i = 1, 2 is about R i The concave function of is easily proved by its second-order derivative being less than 0:
[0049]
[0050] On this basis, the verification scheme for problem (16) is divided into two steps:
[0051] Step 1: Verify whether constraint C0 can be satisfied under constraint C2;
[0052] First calculate like There is no feasible solution; otherwise, if Then there is a feasible solution;
[0053] According to the properties of concave functions, R1[1-Q(f(γ1,n,R1))] has at most two intersections with a given t value; in this case, R1[1-Q(f(γ1,n,R1))]≥t,0≤R1≤R 1,max Through numerical search, the current range of R1 is found to be: in is the lower bound of the feasible R1 at a given t, is the upper bound of feasible R1 at a given t;
[0054] Step 2: If constraint C0 can be satisfied under constraint C2, verify whether constraint C1 can be satisfied under constraint C3;
[0055] At this time, R1 only has a range of values but no fixed value, so C1 is further rewritten as:
[0056] R2[1-Q(f(γ2,n,R2))]≥t / [1-Q(f(γ1,n,R1))] (18)
[0057] Since [1-Q(f(γ1,n,R1))] max =1-ε 1,min =1-Q(f(γ1,n,R 1,min )), if we get the lower bound of R1 Then the constraint range of R2 will be the loosest. If C1 cannot be satisfied under C3 at this time, the larger R1 will constrain R2 more tightly, and C1 will inevitably fail to be satisfied;
[0058] Therefore, if Then there is a feasible solution under this fixed t;
[0059] When the algorithm is initialized, t is a given initial value t (0) , if all constraints can be satisfied under this t, then expand t (1) =2t (0) , until it is found that there is no feasible solution for the first time in a certain cycle, then it can be iterated in a given interval using a search algorithm based on a fixed interval, and finally |t (k+1) -t (k) The t obtained by convergence of |<δ is the maximum value that can be obtained under the problem constraints, that is, the optimal solution to the problem. This algorithm is the best.
[0060] Furthermore, the fixed interval-based search algorithm in step 2.3 is one of a binary search and a golden section search.
[0061] Furthermore, step 2.7 is as follows:
[0062] For this step, the following theorem can be easily proved by contradiction: The optimal t obtained when k Under this condition, R1 has a unique solution, regardless of whether R2 is a unique solution. The optimal t obtained when * The same conclusion can be drawn: R1 is the only solution, so we can search R1 first and then R2 to obtain the optimal transmission rate.
[0063] Therefore, according to the global optimal value t * , by R1[1-Q(f(γ1,n,R1))]≥t *,0≤R1≤R 1,max The numerical search results show that the current range of R1 is: in Yes * The lower bound of the lower feasible R1, is a given t * The upper boundary of the lower feasible R1;
[0064] like Then we need to reduce R1 until we get an equal sign, that is The optimal solution is thus calculated at the same time
[0065] On the contrary, if but Then R2[1-Q(f(γ2,n,R2))]≥t / [1-Q(f(γ1,n,R1))],0≤R2≤R 2,max Through numerical search, the current range of R2 is found to be: Pick To obtain a large T2;
[0066] So far, the optimal short packet transmission rate R = (R1, R2) is obtained.
[0067] The power control method of uplink non-orthogonal multiple access short packet communication of the present invention has the following advantages:
[0068] 1. The present invention adopts global power control based on user fairness, utilizes constraint transformation and feasible solution verification, and provides a power control scheme for uplink non-orthogonal multiple access short packet communication with optimal user fairness through a dual search algorithm, thereby improving the overall performance of the system and ensuring the overall service quality, corresponding to step 2 of claim 2;
[0069] 2. The present invention breaks through the limitation of the prior art of designing the transmission power based on a fixed block error rate, optimizes the uplink transmission power under the constraints of a variable short packet rate and a maximum block error rate, meets the actual needs of the mMTC and uRLLC scenarios in the current Internet of Things, has broad application prospects, and corresponds to claim 3. BRIEF DESCRIPTION OF THE DRAWINGS
[0070] Figure 1 A schematic flow chart of a power control method for uplink non-orthogonal multiple access short packet communication according to the present invention;
[0071] Figure 2 A schematic diagram of an algorithm flow for a base station to estimate the optimal transmission power and short packet transmission rate of each user provided by an embodiment of the present invention;
[0072] Figure 3A power control method for uplink non-orthogonal multiple access short packet communication and a maximum and minimum user effective throughput and transmission power curve of an exhaustive search algorithm provided in an embodiment of the present invention;
[0073] Figure 4 A power control method for uplink non-orthogonal multiple access short packet communication and a calculation time and power / rate search accuracy curve of an exhaustive search algorithm are provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0074] In order to better understand the purpose, structure and function of the present invention, a power control method for uplink non-orthogonal multiple access short packet communication of the present invention is further described in detail below in conjunction with the accompanying drawings.
[0075] refer to Figure 1 , a power control method for uplink non-orthogonal multiple access short packet communication includes: a base station and multiple user equipment, the user equipment accesses the same time-frequency resource block to form a non-orthogonal multiple access system, the base station coordinates the transmission scheme of each user according to the transmission reliability requirements and hardware limitations of the user equipment, and the uplink non-orthogonal multiple access short packet communication process is controlled by a power control scheme based on quality of service and actual constraints. Step 101, before the start of transmission, the user connects to the base station for pairing, and the uplink transmits the non-orthogonal multiple access short packet transmission parameters to be transmitted to the base station; Step 102, the base station receives the user transmission parameters, and estimates the transmission power and short packet transmission rate of each user based on the known channel state information between each user and the base station to maximize the minimum user effective throughput; Step 103, the base station feeds back the estimated transmission power and short packet transmission rate to each user; Step 104, each user transmits a short packet to the base station at the optimal transmission power and transmission rate to complete the uplink non-orthogonal multiple access short packet communication process.
[0076] refer to Figure 2 The embodiment of the present invention provides a typical implementation process in which the base station receives user transmission parameters and estimates the transmission power and short packet transmission rate of each user based on the known channel state information between each user and the base station to maximize the minimum user effective throughput, which specifically includes the following steps:
[0077] Step 2.1, establish the minimax optimization problem and introduce the auxiliary variable t to transform the objective function into a constraint, simplifying the original non-convex optimization problem; the transmission block length n = 500, the power search accuracy of the outer loop α p =0.001, the transmission rate search accuracy of the inner loop α R =0.001 and the iteration termination condition of the inner loop δ=0.001;
[0078] Step 2.2, in the outer loop, set an initial value for the one-dimensional search of the transmit power, which is generally zero;
[0079] Step 2.3: In the inner loop, for a given transmit power, use an iterative algorithm based on binary search to find the optimal value t for the problem at this time. k .
[0080] Step 2.4: Determine the absolute value of the difference between the auxiliary variables t of two adjacent iterations. If it is less than the initial setting value δ of the iteration termination condition, it is determined that the iteration converges to the optimal value and the maximum minimum user effective throughput is obtained, and then go to step 2.5; otherwise, return to step 2.3;
[0081] Step 2.5: In the outer loop, determine the optimal value t obtained in step 2.4 k Is it the global optimum? If it is greater than the existing global optimum value t * , then replace it; otherwise, discard it directly without replacement.
[0082] Step 2.6: Update the transmission power set in the outer loop, and return to step 2.3 until the power search within the feasible range is completed with the initialized power search accuracy.
[0083] Step 2.7: Obtain the global optimal value t * The corresponding optimal transmission power p * , calculate the optimal transmission rate R * .
[0084] Step 2.1 describes the establishment of the minimax optimization problem and the introduction of auxiliary variables t to transform the objective function into constraints, as follows:
[0085] Assume an uplink non-orthogonal multiple access system consisting of a base station and two users. Both the base station and the users have single antennas. The transmission signals of the two users are where s i is the transmitted signal, p i is the transmission power. The signal received by the base station can be expressed as where h i is the channel gain, assuming |h1| 2 >|h2| 2 , w means the mean is zero and the variance is σ 2 Additive white Gaussian noise (AWGN);
[0086] Successive interference cancellation (SIC) is used to separate signals at the base station side. Unlike the decoding order that is always fixed in the downlink non-orthogonal multiple access system, for the uplink system, the decoding order at the base station side is arbitrary because the base station will receive signals from all users. There are many factors that affect the decoding order of the base station, such as channel gain, available resource limitations, and QoS constraints. The present invention sets a decoding order based on channel state information, that is, signals with high channel gain will be decoded preferentially at the base station;
[0087] Therefore, at the base station side, the signal of user 1 is first demodulated, and its signal-to-interference-to-noise ratio is
[0088] The transmission rate R, that is, the amount of effective data sent per second per unit bandwidth, is approximately:
[0089]
[0090] Where γ represents the signal-to-interference-noise ratio, n represents the block length, ε represents the block error rate, C(γ)=log(1+γ) is the Shannon capacity, V(γ)=(1-(1+γ) -2 )(loge) 2 is the channel dispersion, Q -1 (·) is the inverse transform of the Gaussian Q function;
[0091] In a non-orthogonal multiple access short packet communication system, there is a non-negligible error in demodulation, so the probability of correctly demodulating s1 at the base station is (1-ε1), where ε1 = Q(f(γ1,n,R1)), which is obtained by transforming formula (19);
[0092] If s2 is set to be correctly demodulated only after s1 is correctly demodulated during the SIC process, then when the base station demodulates s2, its signal to interference noise ratio is The overall decoding error rate of s2 is expressed as where ε2=Q(f(γ2,n,R2));
[0093] The effective throughput index of uplink non-orthogonal multiple access short packet communication is:
[0094]
[0095] Regarding user fairness, there is a minimax problem, in which power and transmission rate are optimization variables, and the minimum user effective throughput is the objective function. Maximizing the objective function is the optimization direction of the problem, as shown below:
[0096]
[0097] Among them, C0 is the power constraint, C1 and C2 are transmission reliability constraints, and C3 is the practical feasibility constraint;
[0098] because and Therefore, no matter min{T1,T2}=T1 or T2, p1=P max In addition, the following formula can be calculated:
[0099]
[0100] According to formula (22), since ε increases monotonically with R, ε max With R max The corresponding conditions are consistent; therefore, the constraints in problem (21) are transformed to obtain the following problem:
[0101]
[0102] in,
[0103] Solving problem (23) is divided into two stages. The first stage is to find the optimal value under a fixed transmission power p2. The second stage is to perform a one-dimensional search for the transmission power p2, obtain the optimal solution of power p2 by comparing the optimal value obtained in the first stage, and calculate the optimal solutions of transmission rates R1 and R2.
[0104] The issues that need to be addressed in the first phase are as follows:
[0105]
[0106] In order to solve problem (24), an auxiliary variable t is introduced to transform the objective function in the minimax problem into a constraint. The transformed problem is as follows:
[0107]
[0108] Furthermore, in the inner loop described in step 2.3, for a given transmit power, an iterative algorithm based on binary search is used to find the optimal solution to the problem at this time, as follows:
[0109] First, for each fixed value of t, verify whether the problem has a feasible solution, that is, it satisfies all constraints at the same time. There is a theorem: R i [1-Q(f(γ i ,n,R i ))], i = 1, 2 is about R i The concave function of is easily proved by its second-order derivative being less than 0:
[0110]
[0111] On this basis, the verification scheme for problem (25) is divided into two steps:
[0112] Step 1: Verify whether constraint C0 can be satisfied under constraint C2;
[0113] First calculate like There is no feasible solution; otherwise, if Then there is a feasible solution;
[0114] According to the properties of concave functions, R1[1-Q(f(γ1,n,R1))] has at most two intersections with a given t value. In this case, we can use R1[1-Q(f(γ1,n,R1))]≥t,0≤R1≤R 1,max Through numerical search, the current range of R1 is found to be: in is the lower bound of the feasible R1 at a given t, is the upper bound of feasible R1 at a given t;
[0115] Step 2: If C0 can be satisfied under C2, verify whether C1 can be satisfied under C3;
[0116] At this time, R1 only has a range of values but no fixed value, so C1 is further rewritten as:
[0117] R2[1-Q(f(γ2,n,R2))]≥t / [1-Q(f(γ1,n,R1))] (27)
[0118] Since [1-Q(f(γ1,n,R1))] max =1-ε 1,min =1-Q(f(γ1,n,R 1,min )), if we get the lower bound of R1 Then the constraint range of R2 will be the loosest. If C1 cannot be satisfied under C3 at this time, the larger R1 will constrain R2 more tightly, and C1 will inevitably fail to be satisfied;
[0119] Therefore, if Then there is a feasible solution under this fixed t;
[0120] When the algorithm is initialized, t is a given initial value t (0) , if all constraints can be satisfied under this t, then expand t (1) =2t (0) , until it is found that there is no feasible solution for the first time in a certain cycle, then it can be iterated in a given interval by binary search, and finally |t (k+1) -t (k) The t obtained by convergence of |<δ is the maximum value that can be obtained under the problem constraints, that is, the optimal solution to the problem. This algorithm is the best.
[0121] Furthermore, the optimal transmission rate R is obtained by calculating the optimal transmission rate R in step 2.7. * , as follows:
[0122] For this step, the following theorem can be easily proved by contradiction: The optimal t obtained when k Under this condition, R1 has a unique solution, regardless of whether R2 is a unique solution. The optimal t obtained when * The same conclusion can be drawn: R1 is the only solution, so we can search R1 first and then R2 to obtain the optimal transmission rate.
[0123] Therefore, according to the global optimal value t * , by R1[1-Q(f(γ1,n,R1))]≥t * ,0≤R1≤R 1,max The numerical search results show that the current range of R1 is: in Yes * The lower bound of the lower feasible R1, is a given t * The upper boundary of the lower feasible R1;
[0124] if Then we need to reduce R1 until we get an equal sign, that is From this we can calculate the optimal solution at the same time
[0125] On the contrary, if but Then R2[1-Q(f(γ2,n,R2))]≥t / [1-Q(f(γ1,n,R1))],0≤R2≤R 2,max Through numerical search, the current range of R2 is found to be: Pick To get a bigger T2;
[0126] So far, the optimal short packet transmission rate R = (R1, R2) is obtained.
[0127] In summary, the power control method for uplink non-orthogonal multiple access short packet communication of the present invention takes user fairness into consideration and provides a power control scheme for uplink non-orthogonal multiple access short packet communication with optimal user fairness; the uplink transmission power is optimized on the basis of considering the variable short packet transmission rate, and the maximized minimum user effective throughput is obtained through a dual search algorithm, thereby improving the overall performance of the system.
[0128] In this embodiment, h1=0.8, h2=0.2, p max =30dBm,σ 2 =0dBm, ε1=ε2=10 -3 .
[0129] Figure 3They are respectively the maximum and minimum user effective throughput and transmission power curves of a power control method for uplink non-orthogonal multiple access short packet communication of the present invention and an exhaustive search algorithm; as can be seen from the figure, the curve obtained by adopting the power control method for uplink non-orthogonal multiple access short packet communication of the present invention is basically consistent with the curve of the exhaustive search algorithm, which indicates that the power control method for uplink non-orthogonal multiple access short packet communication of the present invention can obtain the optimal power control scheme.
[0130] Figure 4 The power control method for uplink non-orthogonal multiple access short packet communication of the present invention and the calculation time of the exhaustive search algorithm are curves of the power / rate search accuracy; it can be seen from the figure that with the improvement of the search accuracy, the calculation time required for the power control method for uplink non-orthogonal multiple access short packet communication of the present invention increases more slowly than that of the exhaustive search algorithm, which indicates that the power control method for uplink non-orthogonal multiple access short packet communication of the present invention is an effective low-complexity algorithm.
[0131] It is to be understood that the present invention is described by some embodiments, and it is known to those skilled in the art that various changes or equivalent substitutions may be made to these features and embodiments without departing from the spirit and scope of the present invention. In addition, under the teachings of the present invention, these features and embodiments may be modified to adapt to specific circumstances and materials without departing from the spirit and scope of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of this application are within the scope of protection of the present invention.
Claims
1. A power control method for uplink non-orthogonal multiple access short packet communication, characterized in that: The base station coordinates the transmission scheme of each user according to the transmission reliability requirements and hardware limitations of the non-orthogonal multiple access users. The uplink non-orthogonal multiple access short packet communication process is supervised by a power control method based on quality of service and actual constraints; the steps of the power control method are as follows: Step 1: Before the transmission starts, the user connects to the base station for pairing, and transmits the non-orthogonal multiple access short packet transmission parameters to the base station via the uplink; Step 2: The base station receives the non-orthogonal multiple access short packet transmission parameters, and estimates the transmission power and short packet transmission rate of each user based on the known channel state information between each user and the base station to maximize the minimum user effective throughput; Step 3: The base station feeds back the estimated transmission power and short packet transmission rate to each user; Step 4: Each user transmits a short packet to the base station at the optimal transmission power and transmission rate, completing the uplink non-orthogonal multiple access short packet communication process; The non-orthogonal multiple access short packet transmission parameters in step 1 include the maximum allowed block error rate, the short packet transmission block length, and the maximum transmit power upper limit of the user equipment; The step 2 specifically comprises the following steps: Step 2.1, establish a minimax optimization problem according to the non-orthogonal multiple access short packet transmission parameters, introduce auxiliary variables to transform the objective function into constraints, and simplify the original non-convex optimization problem; initialize the block length, the power search accuracy of the outer loop, the transmission rate search accuracy of the inner loop, and the iteration termination condition of the inner loop; Step 2.2, in the outer loop, set the initial value for the one-dimensional search of the transmit power; Step 2.3, in the inner loop, use an iterative algorithm based on fixed interval search to solve the optimal value of the problem under the transmission power set in the outer loop; Step 2.4: Determine the absolute value of the difference between the auxiliary variables of two adjacent iterations. If it is less than the initial setting value of the iteration termination condition, it is determined that the iteration converges to the optimal value and the maximized minimum user effective throughput is obtained, and then proceed to step 2.5; otherwise, return to step 2.3; Step 2.5: In the outer loop, determine whether the optimal value obtained in step 2.4 is the global optimal value. If it is greater than the existing global optimal value, replace it; otherwise, discard it directly without replacement; Step 2.6, update the transmission power set by the outer loop, and return to step 2.3 until the one-dimensional search process is completed for the power within the feasible range with the initialized power search accuracy; Step 2.7, obtain the optimal transmission power corresponding to the global optimal value, and calculate the optimal transmission rate; Step 2.1 describes the establishment of the minimax optimization problem and the introduction of auxiliary variables to transform the objective function into constraints, as follows: Assume an uplink non-orthogonal multiple access system consisting of a base station and two users. Both the base station and the users have single antennas. The transmission signals of the two users are where s i is the transmitted signal, p i is the transmission power; the signal received by the base station is expressed as where h i is the channel gain, set |h1| 2 >|h2| 2 , w means the mean is zero and the variance is σ 2 Additive Gaussian white noise; Successive interference cancellation is used to separate signals at the base station side; the decoding order based on channel state information is set, that is, signals with high channel gain will be decoded first at the base station; Therefore, at the base station side, the signal of user 1 is first demodulated, and its signal-to-interference-to-noise ratio is The transmission rate R, that is, the amount of effective data sent per second per unit bandwidth, is approximately: Where γ represents the signal-to-interference-noise ratio, n represents the block length, ε represents the block error rate, C(γ)=log(1+γ) is the Shannon capacity, V(γ)=(1-(1+γ) -2 )(loge) 2 is the channel dispersion, Q -1 (·) is the inverse transform of the Gaussian Q function; The probability of correctly demodulating s1 at the base station is (1-ε1), where ε1 = Q(f(γ1,n,R1)), which is obtained by transforming formula (1); If s2 is set to be correctly demodulated only after s1 is correctly demodulated during the SIC process, then when the base station demodulates s2, its signal to interference noise ratio is The overall decoding error rate of s2 is expressed as where ε2=Q(f(γ2,n,R2)); The effective throughput index of uplink non-orthogonal multiple access short packet communication is: Regarding user fairness, there is a minimax problem, in which power and transmission rate are optimization variables, and the minimum user effective throughput is the objective function. Maximizing the objective function is the optimization direction of the problem, as shown below: Among them, C0 is the power constraint, C1 and C2 are transmission reliability constraints, and C3 is the practical feasibility constraint; because and Therefore, no matter min{T1,T2}=T1 or T2, p1=P max In addition, the following formula is obtained: According to formula (4), since ε increases monotonically with R, ε max With R max The corresponding conditions are consistent; therefore, the constraints in problem (3) are transformed to obtain the following problem: in, Solving problem (5) is divided into two stages. The first stage is to find the optimal value under the condition of fixed transmission power p2. The second stage is to perform a one-dimensional search for transmission power p2, obtain the optimal solution of power p2 by comparing the optimal value obtained in the first stage, and calculate the optimal solution of transmission rates R1 and R2. The issues that need to be addressed in the first phase are as follows: In order to solve problem (6), an auxiliary variable t is introduced to transform the objective function into a constraint. The transformed problem is as follows: Step 2.3 is as follows: First, for each fixed value of t, verify whether the problem has a feasible solution, that is, it satisfies all constraints at the same time; there is a theorem: R i [1-Q(f(γ i ,n,R i ))], i = 1, 2 is about R i The concave function of is easily proved by its second-order derivative being less than 0: On this basis, the verification scheme for problem (7) is divided into two steps: Step 1: Verify whether constraint C0 can be satisfied under constraint C2; First calculate like There is no feasible solution; otherwise, if Then there is a feasible solution; According to the properties of concave functions, R1[1-Q(f(γ1,n,R1))] has at most two intersections with a given t value; in this case, R1[1-Q(f(γ1,n,R1))]≥t,0≤R1≤R 1,max Through numerical search, the current range of R1 is found to be: in is the lower bound of the feasible R1 at a given t, is the upper bound of feasible R1 at a given t; Step 2: If constraint C0 can be satisfied under constraint C2, verify whether constraint C1 can be satisfied under constraint C3; At this time, R1 only has a range of values but no fixed value, so C1 is further rewritten as: R2[1-Q(f(γ2,n,R2))]≥t / [1-Q(f(γ1,n,R1))](9) Since [1-Q(f(γ1,n,R1))] max =1-ε 1,min =1-Q(f(γ1,n,R 1,min )), if we get the lower bound of R1 Then the constraint range of R2 will be the loosest. If C1 cannot be satisfied under C3 at this time, the larger R1 will constrain R2 more tightly, and C1 will inevitably fail to be satisfied; Therefore, if Then there is a feasible solution under this fixed t; When the algorithm is initialized, t is a given initial value t (0) , if all constraints can be satisfied under this t, then expand t (1) =2t (0) , until it is found that there is no feasible solution for the first time in a certain cycle, then it can be iterated in a given interval using a search algorithm based on a fixed interval, and finally |t (k+1) -t (k) |<δ The t obtained by convergence is the maximum value that can be obtained under the problem constraints, which is the optimal solution to the problem. This algorithm is the best; Step 2.7 is as follows: According to the global optimal value t * , by R1[1-Q(f(γ1,n,R1))]≥t * With 0≤R1≤R 1,max The numerical search results show that the current range of R1 is: in Yes * The lower bound of the lower feasible R1, is a given t * The upper boundary of the lower feasible R1; like Then we need to reduce R1 until we get an equal sign, that is The optimal solution is thus calculated at the same time On the contrary, if but Then R2[1-Q(f(γ2,n,R2))]≥t / [1-Q(f(γ1,n,R1))],0≤R2≤R 2,max Through numerical search, the current range of R2 is found to be: Pick To obtain a large T2; So far, the optimal short packet transmission rate R = (R1, R2) is obtained.
2. The power control method for uplink non-orthogonal multiple access short packet communication according to claim 1, characterized in that: The fixed interval-based search algorithm in step 2.3 is one of a binary search and a golden section search.
