Cell-free network assisted full-duplex enabled industrial internet of things short packet transmission method

Abstract

This invention provides a method for short packet transmission in the Industrial Internet of Things (IIoT) with cellular network-assisted full-duplex capability, comprising: establishing a mathematical model of cellular network-assisted full-duplex; establishing the objective function and constraints of the IIoT short packet transmission problem; reconstructing the problem by decomposing the original problem into a reliability maximization problem and a rate maximization problem; transforming the reliability maximization subproblem into a convex problem and providing a semi-closed solution to the rate maximization subproblem; and solving the transformed problem using an alternating algorithm. This invention can effectively improve the effective throughput of cellular network-assisted full-duplex systems, ensuring low latency and high reliability transmission in the IIoT, and has broad application prospects.

Description

Technical Field

[0001] This invention relates to the field of wireless communication technology, and specifically to a short packet transmission method for industrial IoT that enables full-duplex communication without cellular network assistance. Background Technology

[0002] With the development of sixth-generation (6G) communication systems, the Industrial Internet of Things (IIoT) is envisioned as a new network architecture for realizing smart manufacturing. Utilizing advanced 6G technologies such as Ultra-Reliable and Ultra-Low Latency Communication (URLLC), the IIoT can continuously collect sensor information from various sensors within a factory and rapidly transmit this information to cloud servers for real-time computation and control. In this operating mode, the IIoT can effectively detect and resolve actuator errors, thereby preventing potential industrial accidents. However, machine-to-machine (M2M) communication between sensors, actuators, and controllers has stringent requirements in terms of latency (1ms) and reliability (1-10-9). Furthermore, the realization of industrial automation requires seamless connectivity for a large number of industrial devices. Therefore, simultaneously achieving ultra-high reliability, ultra-low latency, and high connectivity in an IIoT network is a challenging task.

[0003] For low latency, Polyanskiy et al. considered short packet transmission and derived a formula for the maximum achievable data rate under finite block length. The finite block length formula characterizes the complex relationship between data rate, code block length, and decoding error probability. Generally, ultra-reliability, low latency, and high data rate are three contradictory requirements, highly correlated and indispensable, like the three vertices of a triangle. Considering the uplink sensing and downlink control operating modes in the Industrial Internet of Things (IIoT), we propose a Cellular Network-Assisted Full-Duplex (NAFD) architecture to reduce duplex latency. Specifically, a set of access points (APs) are distributed across a cellular network and flexibly divided into transmitting APs and receiving APs. By enabling transmitting and receiving APs to transmit simultaneously on the same frequency, NAFD reduces duplex latency and improves spectral efficiency. Additionally, we also consider using Non-Orthogonal Multiple Access (NOMA) in the IIoT network to increase the number of connected industrial devices and allow multiple devices to share the same time-frequency resources in the power domain to reduce latency. However, both non-cellular network-assisted full-duplex (NOMA) and NOMA reduce the reliability of short packet transmission due to co-channel interference, and the widely used data rate is not suitable as a performance indicator to characterize the level of reliability. Summary of the Invention

[0004] To address the aforementioned issues, this invention provides a short packet transmission method for industrial IoT with full-duplex enablement without cellular network assistance. By jointly optimizing uplink transmission power and downlink beamforming, this invention studies the problem of maximizing effective throughput in a full-duplex system without cellular network assistance while meeting QoS requirements.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A method for short packet transmission in the Industrial Internet of Things (IIoT) that enables full-duplex communication without cellular network assistance includes:

[0007] S1. Establish a mathematical model for full-duplex communication without cellular network assistance;

[0008] S2. Establish the objective function and constraints for the short packet transmission problem in the Industrial Internet of Things (IIoT);

[0009] S3. Problem Restructuring: Decompose the original problem into a reliability maximization problem and a rate maximization problem;

[0010] S4. Transform the reliability maximization subproblem into a convex problem, and give a semi-closed solution to the rate maximization subproblem;

[0011] S5. Use the alternating algorithm to solve the problem transformed from S4.

[0012] Furthermore, step 1 specifically includes the following sub-steps:

[0013] S11. A non-cellular network-assisted full-duplex system is configured with one central processing unit (CPU), H access points (APs) equipped with L antennas, S single-antenna sensors, and K single-antenna actuators. The sensors transmit sensing information of block length N0 to H. U One receiving AP (R-AP), H D Each sending AP (T-AP) has a block length of N. k The control information is given to the actuator, where the CPU knows the channel state information, and the total bandwidth is divided into M orthogonal subcarriers;

[0014] S12. Establish a reliability model for the uplink of a full-duplex system without cellular network assistance. The decoding error probability of uplink transmission is: in The uplink signal-to-interference-plus-noise ratio (SINR) is used for signal-to-interference-plus-noise ratio (SINR). For sensor power, This is the uplink channel gain matrix. The self-interference matrix, Let x be the noise variance, and Q(x) be the Gaussian Q-function. For uplink transmission rate, This is short packet fading, |·| represents absolute value operation, ∑ represents summation operation, and the uplink global error probability is...

[0015] S13. Establish a reliability model for the downlink of a full-duplex system without cellular network assistance, where the downlink transmission decoding error probability is... in This refers to the downlink signal-to-interference-plus-noise ratio (SINR). Let h be the beamforming vector between AP m and actuator k. k，m This is the downlink channel gain matrix. This is due to uplink interference with downlink. Given the downlink transmission rate, the total error probability after the uplink decoding error probability propagates to the downlink is:

[0016] S14. Constructing System Performance Metrics – Effective Throughput, Effective Throughput It is the product of the rate and the decoding error probability.

[0017] Furthermore, step 2 specifically includes the following sub-steps:

[0018] S21. Establish the objective function for maximizing effective throughput: in This means optimizing the sensor transmit power, AP beamforming vector, and uplink / downlink transmission rate respectively to maximize the subsequent function value;

[0019] S22. Establish constraints, including QoS requirements and maximum power consumption constraints for different devices, as follows:

[0020]

[0021]

[0022]

[0023]

[0024]

[0025] in, and This represents the maximum tolerable error probability for both uplink and downlink. and Maximum transmit power for each sensor and AP.

[0026] Furthermore, step 3 specifically includes the following sub-steps:

[0027] S31. Introducing Auxiliary Variables and replace and Reconstruct the data transmission model based on the semidefinite relaxation method;

[0028] S32. Decompose the refactored problem into a reliability maximization problem and a rate maximization problem:

[0029] P1:

[0030]

[0031]

[0032]

[0033]

[0034]

[0035] Where the speed R is given It is a Hermitian matrix, J h =diag([0 (h-1)L I L 0 (H-h)L ]) used to distinguish Hermitian matrices Antennas for each AP;

[0036] P2:

[0037]

[0038]

[0039]

[0040]

[0041]

[0042] Where the power P and the beamforming matrix W are given,

[0043] Furthermore, step 4 specifically includes the following sub-steps:

[0044] S41. Approximate downlink total decoding error probability, simplifying the objective function:

[0045]

[0046] S42. Introducing Auxiliary Variables ρs，m , β k，m Further constraints are imposed on SINR, with the following new constraints:

[0047]

[0048]

[0049]

[0050]

[0051]

[0052]

[0053] Among them, temporary variables and They represent and The first-order Taylor lower bound;

[0054] S43. After transformation, subproblem 1 is approximately a convex problem, with the objective function and constraints as follows:

[0055]

[0056]

[0057]

[0058]

[0059]

[0060]

[0061]

[0062]

[0063]

[0064]

[0065]

[0066]

[0067]

[0068] S44. Based on the concavity / convexity analysis, we can directly obtain the semi-closed solution to subproblem 2:

[0069]

[0070]

[0071] Among them, Γ d and Γ u They are respectively represented as

[0072]

[0073]

[0074] Furthermore, step 5 specifically includes the following sub-steps:

[0075] S51. Initialize all parameters, including the maximum number of iterations I. max Power P is initialized using maximum bit transfer (MRT) precoding. (0) and beamforming vector W (0) Calculate the initial point

[0076] S52. Obtain the optimal solution by solving subproblem 1.

[0077] S53. Substitute the optimal solution of Problem 1 into Subproblem 2, and obtain the optimal uplink and downlink rates based on the closed-form solution of Problem 2. and

[0078] S54. Set i = i + 1;

[0079] S55. Update using the following expression:

[0080]

[0081] S56. Determine if the loop has converged. If the algorithm has converged or has reached the maximum number of iterations I. max Switch to S57, otherwise switch to S52;

[0082] S57. Output the optimal power P for the current cycle. * and beamforming vector W * and update effective throughput

[0083] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0084] 1. This invention considers industrial IoT scenarios enabled by non-cellular network-assisted full-duplex. Non-cellular networks shorten the distance between APs and actuators, and network-assisted full-duplex enables uplink and downlink separation, improving spectrum efficiency and reducing full-duplex latency.

[0085] 2. This invention applies short packet transmission in the uplink and downlink transmission of the Industrial Internet of Things. Short packet transmission can reduce latency, and the joint design of uplink power allocation and downlink precoding reduces the probability of decoding errors caused by short packet transmission.

[0086] Therefore, the present invention can effectively improve the throughput of non-cellular network-assisted full-duplex systems, ensure low latency and high reliability transmission in the Industrial Internet of Things, and has broad application prospects. Attached Figure Description

[0087] Figure 1 The flowchart illustrates the industrial IoT short packet transmission method with full-duplex enablement without cellular network assistance provided by this invention.

[0088] Figure 2 This is a schematic diagram of the alternation algorithm provided by the present invention.

[0089] Figure 3 The performance comparison chart shows the alternation algorithm provided by this invention with those using full-duplex non-orthogonal multiple access (FD-NOMA), full-duplex orthogonal multiple access (FD-OMA), and half-duplex non-orthogonal multiple access (HD-NOMA). Detailed Implementation

[0090] The technical solutions provided by the present invention will be described in detail below with reference to specific embodiments. It should be understood that the following specific embodiments are only used to illustrate the present invention and are not intended to limit the scope of the present invention.

[0091] This invention provides a short packet transmission method for industrial IoT that enables full-duplex communication without cellular network assistance, the process of which is as follows: Figure 1 As shown, it includes the following steps:

[0092] S1. Establish a mathematical model for full-duplex communication without cellular network assistance;

[0093] Specifically, this example establishes a mathematical model for full-duplex without cellular network assistance according to the following steps:

[0094] S11. A non-cellular network-assisted full-duplex system is configured with one central processing unit (CPU), H access points (APs) equipped with L antennas, S single-antenna sensors, and K single-antenna actuators. The sensors transmit sensing information of block length N0 to H. U One receiving AP (R-AP), H D Each sending AP (T-AP) has a block length of N. kThe control information is given to the actuator, where the CPU knows the channel state information, and the total bandwidth is divided into M orthogonal subcarriers;

[0095] S12. Establish a reliability model for the uplink of a full-duplex system without cellular network assistance. The decoding error probability of uplink transmission is: in The uplink signal-to-interference-plus-noise ratio (SINR) is used for signal-to-interference-plus-noise ratio (SINR). For sensor power, This is the uplink channel gain matrix. The self-interference matrix, Let x be the noise variance, and Q(x) be the Gaussian Q-function. For uplink transmission rate, This is short packet fading, |·| represents absolute value operation, ∑ represents summation operation, and the uplink global error probability is...

[0096] S13. Establish a reliability model for the downlink of a cellular network-free full-duplex system, where the downlink transmission decoding error probability is... in This refers to the downlink signal-to-interference-plus-noise ratio (SINORN). Let h be the beamforming vector between AP m and actuator k. k，m This is the downlink channel gain matrix. This is due to uplink interference with downlink. Given the downlink transmission rate, the total error probability after the uplink decoding error probability propagates to the downlink is:

[0097] S14. Constructing System Performance Metrics – Effective Throughput, Effective Throughput It is the product of the rate and the decoding error probability.

[0098] S2. Establish the objective function and constraints for the short packet transmission problem in the Industrial Internet of Things (IIoT);

[0099] Specifically, this example establishes the objective function and constraints for the short packet transmission problem in the Industrial Internet of Things (IIoT) according to the following steps:

[0100] S21. Establish the objective function for maximizing effective throughput: in This means optimizing the sensor transmit power, AP beamforming vector, and uplink / downlink transmission rate respectively to maximize the subsequent function value;

[0101] S22. Establish constraints, including QoS requirements and maximum power consumption constraints for different devices, as follows:

[0102]

[0103]

[0104]

[0105]

[0106]

[0107] in, and This represents the maximum tolerable error probability for both uplink and downlink. and Maximum transmit power for each sensor and AP.

[0108] S3. Problem Restructuring: Decompose the original problem into a reliability maximization problem and a rate maximization problem;

[0109] Specifically, this example breaks down the original problem into two subproblems using the following steps:

[0110] S31. Introducing Auxiliary Variables and replace and Reconstruct the data transmission model based on the semidefinite relaxation method;

[0111] S32. Decompose the refactored problem into a reliability maximization problem and a rate maximization problem:

[0112] P1:

[0113]

[0114]

[0115]

[0116]

[0117]

[0118] Where the speed R is given, It is a Hermitian matrix, J h =diag([0 (h-1)L I L 0 (H-h)L ]) used to distinguish Hermitian matrices Antennas for each AP,

[0119] P2:

[0120]

[0121]

[0122]

[0123]

[0124]

[0125] Where the power P and the beamforming matrix W are given,

[0126] S4. Transform the reliability maximization subproblem into a convex problem, and give a semi-closed solution to the rate maximization subproblem;

[0127] Specifically, this example transforms the reliability maximization subproblem into a convex problem by following these steps, and then provides a semi-closed solution to the rate maximization subproblem:

[0128] S41. Approximate downlink total decoding error probability, simplifying the objective function:

[0129]

[0130] S42. Introducing Auxiliary Variables ρ s，m , β k，m Further constraints are imposed on SINR, with the following new constraints:

[0131]

[0132]

[0133]

[0134]

[0135]

[0136]

[0137] Among them, temporary variables and They represent and The first-order Taylor lower bound;

[0138] S43. After transformation, subproblem 1 is approximately a convex problem, with the objective function and constraints as follows:

[0139]

[0140]

[0141]

[0142]

[0143]

[0144]

[0145]

[0146]

[0147]

[0148]

[0149]

[0150]

[0151]

[0152] S44. Based on the concavity / convexity analysis, we can directly obtain the semi-closed solution to subproblem 2:

[0153]

[0154]

[0155] Among them, Γ d and Γ u They are respectively represented as

[0156]

[0157]

[0158] S5. Use the alternating algorithm to solve the problem transformed from S4.

[0159] Specifically, this example uses an alternating algorithm to solve the problem after S4 transformation, following these steps:

[0160] S51. Initialize all parameters, including the maximum number of iterations I. max Power P is initialized using maximum bit transfer (MRT) precoding. (0) and beamforming vector W (0) Calculate the initial point

[0161] S52. Obtain the optimal solution by solving subproblem 1.

[0162] S53. Substitute the optimal solution of Problem 1 into Subproblem 2, and obtain the optimal uplink and downlink rates based on the closed-form solution of Problem 2. and

[0163] S54. Set i = i + 1;

[0164] S55. Update using the following expression:

[0165]

[0166] S56. Determine if the loop has converged. If the algorithm has converged or has reached the maximum number of iterations I. max Switch to S57, otherwise switch to S52;

[0167] S57. Output the optimal power P for the current cycle. * and beamforming vector W * and update effective throughput

[0168] In summary, this invention provides a method for short packet transmission in the Industrial Internet of Things (IIoT) with cellular network-assisted full-duplex capability, comprising: establishing a mathematical model of cellular network-assisted full-duplex; establishing the objective function and constraints of the IIoT short packet transmission problem; reconstructing the problem by decomposing the original problem into a reliability maximization problem and a rate maximization problem; transforming the reliability maximization subproblem into a convex problem and providing a semi-closed solution to the rate maximization subproblem; and solving the transformed problem using an alternating algorithm. This invention can effectively improve the effective throughput of cellular network-assisted full-duplex systems, ensuring low latency and high reliability transmission in the IIoT, and has broad application prospects.

[0169] The technical means disclosed in this invention are not limited to those disclosed in the above embodiments, but also include technical solutions composed of any combination of the above technical features. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of this invention, and these improvements and modifications are also considered within the scope of protection of this invention.

Claims

1. A short packet transmission method for industrial IoT with full-duplex enablement without cellular network assistance, characterized in that, include: S1. Establish a mathematical model for full-duplex operation without cellular network assistance; S2. Based on the mathematical model, establish the objective function and constraints for the short packet transmission problem in the Industrial Internet of Things (IIoT). S3. Reconstruct the short packet transmission problem in the Industrial Internet of Things, decomposing the original problem into a reliability maximization problem and a rate maximization problem; S4. Transform the reliability maximization subproblem into a convex problem, and give a semi-closed solution to the rate maximization subproblem; S5. Use an alternating algorithm to solve the reliability maximization problem and the rate maximization problem after the transformation in S4; Step S1 specifically includes the following sub-steps: S11. Configured with one central processing unit. Each equipped with The connection point of the root antenna. A single antenna sensor and A cellular-free, full-duplex system with a single-antenna actuator, where the sensor transmitter block length is... Perceived information to One receiving AP, The sending block length of each AP is... The control information is given to the actuator, where the central processing unit knows the channel state information, and the total bandwidth is divided into... Orthogonal subcarriers; S12. Establish a reliability model for the uplink of a full-duplex system without cellular network assistance, where the decoding error probability of uplink transmission is... ,in The uplink signal-to-interference-plus-noise ratio (SINR) is used for signal-to-interference-plus-noise ratio (SINR). For sensor power, This is the uplink channel gain matrix. The self-interference matrix, For noise variance, It is a Gaussian Q-function. , For uplink transmission rate, It is short packet fading. This represents the absolute value operation. This represents a summation operation, with an uplink global error probability of . ; S13. Establish a reliability model for downlink in a full-duplex system without cellular network assistance, where the downlink transmission decoding error probability is... ,in This refers to the downlink signal-to-interference-plus-noise ratio (SINR). For AP With actuator Beamforming vectors between This is the downlink channel gain matrix. This is due to uplink interference with downlink. Given the downlink transmission rate, the total error probability after the uplink decoding error probability propagates to the downlink is: ; S14. Constructing system performance metrics – effective throughput, effective throughput It is the product of the rate and the decoding error probability; Step S2 specifically includes the following sub-steps: S21. Establish the objective function for maximizing effective throughput: ,in This means optimizing the sensor transmit power, AP beamforming vector, and uplink / downlink transmission rate respectively to maximize the subsequent function value; S22. Establish constraints, including QoS requirements and maximum power consumption constraints for different devices, as follows: ； ； ； ； ； in, and This represents the maximum tolerable error probability for both uplink and downlink. and Maximum transmit power for each sensor and AP; Step S3 specifically includes the following sub-steps: S31. Introducing Auxiliary Variables and replace and The data transmission model is reconstructed based on the semidefinite relaxation method; S32. Decompose the reconstructed problem into a reliability maximization problem P1 and a rate maximization problem P2: ； Among them, rate Given, It is a Hermitian matrix. Used to distinguish Hermitian matrices Antennas for each AP; ； Among them, power and beamforming matrix Given, , ; Step S4 specifically includes the following sub-steps: S41. Approximate the total downlink decoding error probability, simplifying the objective function: ； S42. Introducing Auxiliary Variables , , , Further constraints are imposed on SINR, with the following new constraints: ； ； Among them, temporary variables and They represent and The first-order Taylor lower bound; S43. After transformation, the reliability maximization problem P1 is approximately a convex problem, with the objective function and constraints as follows: ； S44. Based on the concavity / convexity analysis, the semi-closed solution to the rate maximization problem P2 is obtained: ； in, and They are respectively represented as 。 2. The method for short packet transmission in industrial IoT with full-duplex enablement without cellular network assistance as described in claim 1, characterized in that, Step S5 specifically includes the following sub-steps: S51. Initialize all parameters, including the maximum number of iterations. Power is initialized using maximum bit transmission precoding. and beamforming vector Calculate the initial point , , , , , ; S52. Obtain the optimal solution by solving the transformed reliability maximization problem P1. , , , ; S53. Substitute the optimal solution of problem P1 into problem P2, and obtain the optimal uplink and downlink rates based on the closed-form solution of problem P2. and ; S54. Settings S55. Update using the following expression: , , ； S56. Determine if the loop has converged. If the algorithm converges or reaches the maximum number of iterations... Switch to S57, otherwise switch to S52; S57. Output the optimal power for the current cycle. and beamforming vector and update effective throughput .

Images