Optimization method for integrated noma communication and over-the-air computation system

Abstract

The present invention provides an optimization method for an integrated NOMA communication and over-the-air computation system. The method comprises: constructing an uplink channel model on the basis of an integrated NOMA communication and over-the-air computation system; constructing an optimization problem model on the basis of a plurality of preset constraint conditions and the uplink channel model, wherein the preset constraint conditions comprise minimizing the mean square error sum of over-the-air computation performed by the uplink channel model, the achievable information rate sum of NOMA communication performed by the uplink channel model being greater than a preset value, and the signal transmission power of the uplink channel model being within a preset range; and solving the optimization problem model to obtain a control input of the integrated NOMA communication and over-the-air computation system, and controlling, on the basis of the control input, the operation of the integrated NOMA communication and over-the-air computation system. The solution can implement joint deployment of NOMA communication and over-the-air computation while ensuring signal transmission performance and computational accuracy.

Description

Optimization Methods for NOMA Integrated Communication and In-flight Computing System

[0001] This disclosure claims priority to patent application No. 202411015983.6, filed on July 26, 2024, entitled "Optimization Method for NOMA Integrated Communication and In-flight Computing System," the entire contents of which are incorporated herein by reference. Technical Field

[0002] This disclosure relates to the field of data communication technology, and more specifically, to an optimization method, optimization device, computer-readable storage medium, and NOMA communication and airborne computing integrated system. Background Technology

[0003] NOMA communication allows multiple wireless communication links to simultaneously occupy the same communication resource block, effectively improving spectrum utilization. The main characteristic of NOMA communication is its simultaneous support for multiple device terminals. However, the receiving device needs to use complex successive interference cancellation (SIC) technology to decode information from weaker users by stronger ones, eliminating interference caused by the superposition of multiple users. Power-domain NOMA allocates different transmit powers based on the differences in channel quality between different users. In this case, multiple users can share the same time-frequency coding resources. Code-domain NOMA is similar to CDMA or multi-carrier CDMA and can be implemented through spread spectrum sequences. In uplink NOMA networks, multiple transmitters transmit signals simultaneously, and the maximum transmit power of each transmitter is limited. Joint coding between transmitters is not possible, but the receiving end can decode the requested signal through SIC. Uplink NOMA networks can accommodate a large number of users. Due to sufficient differences in channel gain between different terminals and the base station, when decoding a specific terminal signal, the receiving base station treats the channel gain of other terminals as interference. The base station can first decode the signal of the terminal with good channel quality and then cancel the decoded signal from the superimposed signal, and so on, until all signals are decoded.

[0004] Over-the-air computing leverages the waveform superposition characteristics of multiple access channels to treat the wireless propagation medium as an equivalent calculator, enabling data computation and information communication for massive distributed devices. This makes low-latency wireless data aggregation in large-scale IoT networks possible. Both over-the-air computing and NOMA communication utilize the superposition properties of access channels, and their similar technical attributes make the integration of communication and computing possible, but related research is still very lacking.

[0005] However, in some solutions known to the inventors, the network involves the coupling of communication and computing scenarios, requiring the communication system to simultaneously meet the needs of both scenarios. However, in some solutions known to the inventors, the network architecture merely deploys the two scenarios separately, leading to mutual interference between the two devices. The inventors realized that this seriously affects the system's communication quality and computing accuracy.

[0006] Summary of the Invention

[0007] The main objective of this disclosure is to provide an optimization method, optimization device, computer-readable storage medium, and integrated NOMA communication and in-flight computing system, so as to at least solve the problem that the joint deployment of NOMA communication and in-flight computing cannot guarantee signal transmission performance and computing accuracy.

[0008] To achieve the above objectives, according to one aspect of the disclosure, an optimization method for an integrated NOMA communication and over-the-air computing system is provided. The integrated NOMA communication and over-the-air computing system includes a communication base station, a data fusion center, and multiple terminal devices. The method includes: constructing an uplink channel model based on the integrated NOMA communication and over-the-air computing system, the uplink channel model being used to simulate signal changes during the process of the terminal devices transmitting signals to the communication base station and the data fusion center; constructing an optimization problem model based on multiple preset constraints and the uplink channel model, the preset constraints including minimizing the mean square error of the uplink channel model for over-the-air computing, achieving an achievable information rate greater than a preset value for the uplink channel model for NOMA communication, and ensuring the signal transmission power of the uplink channel model is within a preset range; solving the optimization problem model to obtain the control input of the integrated NOMA communication and over-the-air computing system, and controlling the operation of the integrated NOMA communication and over-the-air computing system according to the control input, the control input including at least the signal transmission power and a noise reduction factor.

[0009] Optionally, constructing an uplink channel model based on the NOMA communication and over-the-air computing integrated system includes: simulating the process of the terminal device encoding the data to be transmitted to obtain a first target signal, and obtaining a first target formula; simulating the process of the first target signal being transmitted through a first target channel to obtain a second target signal, and obtaining a second target formula; simulating the process of the data fusion center denoising the second target signal and performing over-the-air computing to obtain the calculation result, and obtaining a third target formula; constructing a mean square error calculation formula for the over-the-air computing performed by the data fusion center, and obtaining a fourth target formula, where the first target channel is the uplink channel for data transmission from the terminal device to the data fusion center; simulating the process of the first target signal being transmitted through a second target channel to obtain a third target signal, and obtaining a fifth target formula; constructing an information reachability rate calculation formula for NOMA communication performed by the communication base station, and obtaining a sixth target formula, where the second target channel is the uplink channel for data transmission from the terminal device to the communication base station; and constructing the uplink channel model based on the first target formula, the second target formula, the third target formula, the fourth target formula, the fifth target formula, and the sixth target formula.

[0010] Optionally, the process of encoding the data to be transmitted by the terminal device to obtain the first target signal is simulated to obtain the first target formula, including: acquiring the data to be transmitted and the first target signal, and constructing the first target formula based on the data to be transmitted and the first target signal. Where, x k The first target signal, h is the conjugate of the first target channel. k,c For the first target channel, |h k,c | represents the modulus of the first target channel, s k The data to be sent.

[0011] Optionally, the process of transmitting the first target signal through the first target channel to obtain the second target signal is simulated to obtain a second target formula. The process of denoising the second target signal and performing over-the-air calculations based on the data fusion center is simulated to obtain a third target formula. A mean square error calculation formula for over-the-air calculations by the data fusion center is constructed to obtain a fourth target formula, including: acquiring the first target signal and the second target signal, and constructing the second target formula based on the first target signal and the second target signal. Among them, y c p is the second target signal. k The signal transmission power is n, K is the number of terminal devices, and n is the number of terminal devices. cThe received ambient noise of the data fusion center; the calculation result is obtained, and the third target formula is constructed based on the calculation result and the second target signal: in, The estimated value of the calculation result is η, and the denoising factor is η. The mean square error calculation formula for the data fusion center's aerial computation is constructed to obtain the fourth objective formula: Where s is the actual value of the calculation result, and s satisfies When s is unknown, CMSE satisfies in For n c The noise variance.

[0012] Optionally, the process of transmitting the first target signal through the second target channel to obtain the third target signal is simulated to obtain the fifth target formula. A formula for calculating the reachable information rate of the communication base station performing NOMA communication is then constructed to obtain the sixth target formula, which includes: acquiring the first target signal and the third target signal, and constructing the fifth target formula based on the first target signal and the third target signal. Among them, y b For the third target signal, h k,b For the second target channel, n b The received ambient noise of the communication base station; the formula for calculating the achievable information rate of the communication base station for NOMA communication is constructed to obtain the sixth objective formula: R k =log2(1+γ) k ); where R k γ is the rate at which the information can be reached. k The signal-to-dryness ratio of the data received by the communication base station, |h k,b | represents the modulus of the second target channel. For n b The noise variance, |h k,b |satisfy|h 1,b |≤|h 2,b |≤...≤|h K,b |

[0013] Optionally, constructing an optimization problem model based on multiple preset constraints and the uplink channel model includes: constructing the optimization problem model based on the uplink model: Where r is the preset value.

[0014] Optionally, solving the optimization problem model to obtain the control input of the NOMA communication and over-the-air computing integrated system includes: an acquisition step, acquiring a preset denoising factor, and transforming the optimization problem model according to the preset denoising factor. Where τ=2 r -1; Construction steps: Construct the Lagrangian function based on the transformed optimization problem model: Where α is a Lagrange multiplier; the first calculation step is to update the signal transmission power based on the Lagrange function: The second calculation step involves updating the preset noise reduction factor based on the updated signal transmission power: The first calculation step or the second calculation step is repeated at least once until the number of iterations reaches a preset threshold. The updated signal transmission power and the updated preset denoising factor are then determined as the control input.

[0015] According to another aspect of this disclosure, an optimization apparatus for a NOMA (Noise, MA, and Telecommunications) integrated communication and over-the-air computing system is provided. The NOMA integrated communication and over-the-air computing system includes a communication base station, a data fusion center, and multiple terminal devices. The apparatus includes: a first construction unit configured to construct an uplink channel model based on the NOMA integrated communication and over-the-air computing system, the uplink channel model being used to simulate signal changes during the process of the terminal devices transmitting signals to the communication base station and the data fusion center; a second construction unit configured to construct an optimization problem model based on multiple preset constraints and the uplink channel model, the preset constraints including minimizing the mean square error of the uplink channel model for over-the-air computing, achieving an achievable information rate greater than a preset value for the uplink channel model for NOMA communication, and ensuring the signal transmission power of the uplink channel model is within a preset range; and a calculation unit configured to solve the optimization problem model to obtain the control input of the NOMA integrated communication and over-the-air computing system, and control the operation of the NOMA integrated communication and over-the-air computing system according to the control input, the control input including at least the signal transmission power and a noise reduction factor.

[0016] According to another aspect of this disclosure, a computer-readable storage medium is provided, the computer-readable storage medium including a stored program, wherein, when the program is executed, it controls the device on which the computer-readable storage medium is located to perform any of the methods described.

[0017] According to another aspect of this disclosure, a NOMA communication and over-the-air computing integrated system is provided, comprising: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including methods for any one of the methods.

[0018] Applying the technical solution of this disclosure, in the optimization method of the aforementioned NOMA communication and over-the-air computing integrated system, firstly, an uplink channel model is constructed based on the aforementioned NOMA communication and over-the-air computing integrated system. The uplink channel model is used to simulate the signal changes during the process of the terminal device transmitting signals to the communication base station and the data fusion center. Then, an optimization problem model is constructed based on multiple preset constraints and the aforementioned uplink channel model. The preset constraints include minimizing the mean square error of the uplink channel model for over-the-air computing, ensuring that the achievable information rate of the uplink channel model for NOMA communication is greater than a preset value, and ensuring that the signal transmission power of the uplink channel model is within a preset range. Finally, the optimization problem model is solved to obtain the control input of the aforementioned NOMA communication and over-the-air computing integrated system, and the operation of the aforementioned NOMA communication and over-the-air computing integrated system is controlled according to the aforementioned control input. The aforementioned control input includes at least the aforementioned signal transmission power and a noise reduction factor. This application establishes an uplink channel model based on an integrated wireless NOMA communication and over-the-air computing system. Based on this uplink channel model, a data model (i.e., an optimization problem model) is built to minimize the computational error of terminal devices. Optimization is achieved by designing signal encoding for IoT devices to jointly optimize signal transmission power and denoising factors. Under the constraints of maximizing terminal device transmission power and the system's achievable information rate and signal transmission power, the mean square error of the system's computation is minimized. Control parameters for the integrated NOMA communication and over-the-air computing system are determined, and the system's operation is controlled according to these parameters. This method addresses the problem of unreliable signal transmission performance and computational accuracy in the joint deployment of NOMA communication and over-the-air computing. Attached Figure Description

[0019] Figure 1 shows a hardware structure block diagram of an optimized mobile terminal for a NOMA communication and over-the-air computing integrated system provided in an embodiment of this application.

[0020] Figure 2 shows a flowchart illustrating an optimization method for an integrated NOMA communication and over-the-air computing system according to an embodiment of this application.

[0021] Figure 3 illustrates a communication interaction diagram of an integrated NOMA communication and over-the-air computing system according to an embodiment of this application;

[0022] Figure 4 shows a structural block diagram of an optimization device for an integrated NOMA communication and over-the-air computing system provided according to an embodiment of this application.

[0023] The above figures include the following reference numerals:

[0024] 102. Processor; 104. Memory; 106. Transmission device; 108. Input / output device. Detailed Implementation

[0025] It should be noted that, unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other. This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0026] To enable those skilled in the art to better understand the present disclosure, the technical solutions of the present disclosure will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of the present disclosure, and not all embodiments. Based on the embodiments of the present disclosure, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present disclosure.

[0027] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate for the embodiments of this disclosure described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0028] Definitions:

[0029] NOMA (Non-Orthogonal Multiple Access): A non-orthogonal multiple access technology, also known as non-orthogonal multiple access communication (NOMA). In NOMA communication, multiple users can share the same spectrum resources and achieve data transmission through different power levels and coding schemes. This technology can significantly improve spectrum efficiency and system capacity, and is suitable for 5G and future communication systems. NOMA communication has become one of the research hotspots in the field of wireless communication, attracting widespread attention and research.

[0030] CDMA (Code Division Multiple Access) is a digital communication technology that allows multiple users to communicate simultaneously on the same frequency band, distinguishing them by using different pseudo-random codes. CDMA technology is widely used in mobile communication systems, such as 3G and 4G networks.

[0031] As described in the background section, in some solutions known to the inventors, the network involves the coupling of communication and computing scenarios, and the communication system needs to simultaneously meet the requirements of both scenarios. However, in some solutions known to the inventors, the network architecture simply deploys the two scenarios separately, which leads to mutual interference between the two devices, severely affecting the communication quality and computing accuracy of the system. To address the problem of unreliable signal transmission performance and computing accuracy in the joint deployment of NOMA communication and over-the-air computing, embodiments of this disclosure provide an optimization method, optimization device, computer-readable storage medium, and an integrated NOMA communication and over-the-air computing system.

[0032] The technical solutions in the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings.

[0033] The methods and embodiments provided in this disclosure can be executed in a mobile terminal, computer terminal, or similar computing device. Taking a mobile terminal as an example, FIG1 is a hardware structure block diagram of a mobile terminal for an optimization method of an integrated NOMA communication and over-the-air computing system according to an embodiment of this disclosure. As shown in FIG1, the mobile terminal may include one or more (only one is shown in FIG1) processors 102 (processor 102 may include, but is not limited to, processing devices such as microprocessors MCUs or programmable logic devices FPGAs) and a memory 104 for storing data. The mobile terminal may also include a transmission device 106 for communication functions and an input / output device 108. It will be understood by those skilled in the art that the structure shown in FIG1 is only illustrative and does not limit the structure of the mobile terminal. For example, the mobile terminal may also include more or fewer components than shown in FIG1, or have a different configuration than shown in FIG1.

[0034] The memory 104 can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to the device information display method in this embodiment. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, thus implementing the above-described method. The memory 104 may include high-speed random access memory and non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the mobile terminal via a network. Examples of the aforementioned networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof. The transmission device 106 is used to receive or send data via a network. Specific examples of the aforementioned networks may include wireless networks provided by the mobile terminal's communication provider. In one example, the transmission device 106 includes a network interface controller (NIC), which can be connected to other network devices via a base station to communicate with the Internet. In one example, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0035] This embodiment provides an optimization method for a NOMA communication and over-the-air computing integrated system running on a mobile terminal, computer terminal, or similar computing device. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0036] Figure 2 is a flowchart of an optimization method for a NOMA communication and over-the-air computing integrated system according to an embodiment of the present disclosure. As shown in Figure 2, the method includes the following steps:

[0037] Step S201: Construct an uplink channel model based on the above-mentioned NOMA communication and over-the-air computing integrated system. The uplink channel model is used to simulate the changes in signals during the process of the terminal device sending signals to the communication base station and the data fusion center.

[0038] In some embodiments, this disclosure provides an uplink channel model that integrates wireless NOMA (non-orthogonal multiple access) communication and over-the-air computing. The uplink channel model is configured to simulate the changes in signals during the process of the terminal sending signals to the communication base station for NOMA communication and the terminal sending signals to the data fusion center for over-the-air computing.

[0039] In some embodiments, as shown in Figure 3, in the above-described integrated wireless NOMA communication and over-the-air computing system, any user can simultaneously interact with both the wireless communication base station and the data fusion center.

[0040] Step S202: Construct an optimization problem model based on multiple preset constraints and the above uplink channel model. The preset constraints include minimizing the mean square error of the above uplink channel model for in-flight calculations, ensuring that the achievable information rate of the uplink channel model for NOMA communication is greater than a preset value, and ensuring that the signal transmission power of the above uplink model is within a preset range.

[0041] In some embodiments, this disclosure sets an optimization objective based on the above-described uplink channel model, namely, minimizing the above-described system air computation error, and constructs a mathematical model for the objective to obtain the above-described optimization problem model.

[0042] Step S203: Solve the above optimization problem model to obtain the control input of the above NOMA communication and airborne computing integrated system, and control the operation of the above NOMA communication and airborne computing integrated system according to the above control input. The above control input includes at least the above signal transmission power and noise reduction factor.

[0043] In some embodiments, by designing the signal encoding of the terminal device and jointly optimizing the transmission power and denoising factor of the terminal device, the above optimization problem model is solved to minimize the mean square error of the system's calculation under the constraints of the maximum transmission power of the terminal device, the system, and the rate, so as to meet the requirements of system computing power and system latency.

[0044] In this embodiment, firstly, an uplink channel model is constructed based on the aforementioned NOMA communication and over-the-air computing integrated system. This uplink channel model is used to simulate signal changes during the process of the terminal device transmitting signals to the communication base station and the data fusion center. Then, an optimization problem model is constructed based on multiple preset constraints and the aforementioned uplink channel model. The preset constraints include minimizing the mean square error of the uplink channel model performing over-the-air computing, ensuring that the achievable information rate of the uplink channel model performing NOMA communication is greater than a preset value, and ensuring that the signal transmission power of the uplink channel model is within a preset range. Finally, the optimization problem model is solved to obtain the control input of the aforementioned NOMA communication and over-the-air computing integrated system, and the operation of the aforementioned NOMA communication and over-the-air computing integrated system is controlled according to the aforementioned control input. The aforementioned control input includes at least the aforementioned signal transmission power and a noise reduction factor. This disclosure establishes an uplink channel model based on an integrated wireless NOMA communication and over-the-air computing system. Based on this uplink channel model, a data model (i.e., an optimization problem model) is established to minimize the computational error of terminal devices. Optimization is achieved by designing signal encoding for IoT devices to jointly optimize signal transmission power and denoising factors. Under the constraints of maximizing terminal device transmission power and the system's achievable information rate and signal transmission power, the mean square error of the system's computation is minimized. Control parameters for the integrated NOMA communication and over-the-air computing system are determined, and the system's operation is controlled according to these parameters. This method addresses the problem of unreliable signal transmission performance and computational accuracy in the joint deployment of NOMA communication and over-the-air computing.

[0045] To construct the aforementioned uplink channel model, in one optional implementation, step S201 includes:

[0046] Step S2011: Simulate the process of the terminal device encoding the data to be transmitted to obtain the first target signal, and obtain the first target formula;

[0047] In some embodiments, it is assumed that the integrated wireless NOMA communication and over-the-air computing system deploys a communication base station and a data fusion center, as well as K single-antenna IoT devices simultaneously performing uplink data transmission and over-the-air computing. Based on this, according to the signal encoding of the data to be transmitted by the terminal device during data transmission, the aforementioned first objective formula is constructed to characterize the changes in data during the signal encoding process.

[0048] Step S2012: Simulate the process of the first target signal being transmitted through the first target channel to obtain the second target signal, and obtain the second target formula. Simulate the process of the data fusion center denoising the second target signal and performing over-the-air calculation to obtain the calculation result, and obtain the third target formula. Construct the mean square error calculation formula for the over-the-air calculation performed by the data fusion center, and obtain the fourth target formula. The first target channel is the uplink channel for the terminal device to transmit data to the data fusion center.

[0049] In some embodiments, the second target formula described above is constructed to characterize the signal changes during the transmission of the first target signal by the first target channel. Then, based on the process of the data fusion center using superimposed data from the air interface channel and performing denoising processing on the acquired signal, the third target formula is constructed. Finally, the error calculation formula for the accuracy of in-flight computation is constructed to obtain the fourth target formula.

[0050] Step S2013: Simulate the process of the first target signal being transmitted through the second target channel to obtain the third target signal, and obtain the fifth target formula. Construct the information reachability rate calculation formula for the communication base station performing NOMA communication, and obtain the sixth target formula. The second target channel is the uplink channel for the terminal device to transmit data to the communication base station.

[0051] In some embodiments, the fifth target formula is constructed to characterize the signal changes during the transmission of the first target signal by the second target channel, and then the sixth target formula is obtained based on the calculation formula of the signal reachability rate of each terminal signal obtained by the communication base station using SIC basic decoding for NOMA communication.

[0052] Step S2014: Construct the uplink channel model based on the first objective formula, the second objective formula, the third objective formula, the fourth objective formula, the fifth objective formula, and the sixth objective formula.

[0053] In some embodiments, a channel model is constructed based on the first objective formula, the second objective formula, the third objective formula, the fourth objective formula, the fifth objective formula, and the sixth objective formula to obtain the uplink channel model.

[0054] In order to construct the first target formula described above, in one optional implementation, step S2011 includes:

[0055] Step S20111: Obtain the data to be transmitted and the first target signal, and construct the first target formula based on the data to be transmitted and the first target signal:

[0056] Where, x k The first target signal mentioned above, h is the conjugate of the first target channel mentioned above. k,c For the first target channel mentioned above, |h k,c | represents the modulus of the first target channel mentioned above, s k The above is the data to be sent.

[0057] In some embodiments, the data fusion center operates in time-division duplex mode. Before wireless uplink transmission, the data fusion center broadcasts pilot signals to all terminals. Terminal k uses the reciprocity of the wireless channel to estimate the channel h from terminal k to the data fusion center. k,c Terminal k according to channel h k,c The data is encoded to obtain the first target signal mentioned above, and the corresponding formula for the first target is as shown above.

[0058] It is understood that in this disclosure, the single-antenna IoT device sends only one transmission signal each time, which is sent to the communication base station and the data fusion center for signal transmission and over-the-air computing. Therefore, in the process of signal encoding, signal encoding can be performed only according to the data fusion center.

[0059] In order to construct the above-mentioned second objective formula, third objective formula, and fourth objective formula, in an optional implementation, step S2012 includes:

[0060] Step S20121: Obtain the first target signal and the second target signal, and construct the second target formula based on the first target signal and the second target signal.

[0061] Among them, y c For the second target signal mentioned above, p k Where n is the signal transmission power, K is the number of terminal devices, and n is the number of terminal devices. c The ambient noise received by the aforementioned data fusion center;

[0062] In some embodiments, the second target formula is constructed based on the influence of the channel on the first target signal during the transmission of the first target signal to the data fusion center via the first target channel.

[0063] Step S20122: Obtain the above calculation results, and construct the above third target formula based on the above calculation results and the above second target signal:

[0064] in, Here is an estimate of the above calculation results, and η is the above denoising factor;

[0065] In some embodiments, the data fusion center directly uses the superposition property of the air interface channel to denoise the obtained signal, directly obtaining the calculation results of K terminal devices. The ideal air interface calculation result is: However, in actual calculations, the aforementioned first target signal is unknown to the data fusion center, which only receives the second target signal. Therefore, this disclosure sets up an estimate based on the received second target signal and the denoising factor.

[0066] Step S20123: Construct the mean square error calculation formula for the above data fusion center to perform aerial calculations, and obtain the fourth objective formula:

[0067] Where s is the actual value of the above calculation result, and s satisfies When s is unknown, CMSE satisfies in For n c The noise variance.

[0068] In some embodiments, this disclosure constructs a formula for calculating the mean squared error (MSE) of an over-the-air computation in a data fusion center, using the MSE as a criterion for evaluating simulation results. Furthermore, the MSE is defined as shown in the formula above. Since the ideal over-the-air computation result is unknown, this disclosure sets a formula based on... The system's mean square error is calculated. This mean square error is used to assess the performance of the uplink channel model and to correct it.

[0069] In order to construct the fifth objective formula and the sixth objective formula mentioned above, in an optional implementation, step S2013 includes:

[0070] Step S20131: Obtain the first target signal and the third target signal, and construct the fifth target formula based on the first target signal and the third target signal.

[0071] Among them, y b For the third target signal mentioned above, h k,b For the second target channel mentioned above, n b The ambient noise received by the aforementioned communication base stations;

[0072] In some embodiments, the fifth target formula is constructed based on the influence of the channel on the first target signal during the transmission of the first target signal to the communication base station via the second target channel.

[0073] Step S20132, construct the information reachability rate calculation formula for the above communication base station to obtain the sixth target formula: Rk =log2(1+γ) k );

[0074] Among them, R k For the rate at which the above information can be obtained, γ k The signal-to-dryness ratio of the data received by the aforementioned communication base station, |h k,b | represents the modulus of the second target channel mentioned above. For n b The noise variance, |h k,b |satisfy|h 1,b |≤|h 2,b |≤...≤|h K,b |

[0075] In some embodiments, the communication base station uses walk-through interference cancellation technology to obtain the signal of each terminal for NOMA communication, and assumes that the channel gains are arranged in ascending order, i.e., |h 1,b |≤|h 2,b |≤...≤|h K,b The signal-to-drying ratio of the first K-1 devices satisfies:

[0076] The signal-to-dryness ratio of the Kth device satisfies: Then, the signal-to-dryness ratio of the terminal device k is substituted into the formula: R k =log2(1+γ) k It can calculate the achievable information rate of the k-th device.

[0077] In order to construct a mathematical model, in one optional implementation, step S202 above includes:

[0078] Step S2021: Construct the optimization problem model based on the uplink channel model described above.

[0079] Where r is the aforementioned preset value.

[0080] In some embodiments, the CMSE and NOMA communications transmitted by each terminal in the system via pilot air calculations and the rate are both related to the terminal's transmit power p. k The CMSE is related to the noise reduction factor. Based on the CMSE calculation formula and the achievable information rate calculation formula, an optimization problem model is established, resulting in the model described above. The formulas above represent the problem of minimizing the mean square error of in-flight computation under constraints, system and rate constraints, and transmit power constraints for each terminal, respectively.

[0081] In order to solve the above optimization problem model to obtain the control parameters, in an optional implementation, step S203 includes:

[0082] Step S2031, obtaining the preset denoising factor, and transforming the above optimization problem model according to the preset denoising factor:

[0083] Where τ=2 r -1;

[0084] In some embodiments, the optimization problem model described above can be solved using a low-complexity local optimum method. Given a denoising factor, i.e., the aforementioned preset denoising factor, the solution to the sub-problem of terminal transmit power can be transformed into the aforementioned model.

[0085] Step S2032, Construction Step: Construct the Lagrangian function based on the transformed optimization problem model described above.

[0086] Where α is a Lagrange multiplier;

[0087] In some embodiments, the corresponding Lagrangian function, based on the transformed model described above, is represented as shown in the above equation.

[0088] Step S2033, the first calculation step, updates the above signal transmission power according to the Lagrange function:

[0089] In some embodiments, the optimal power closed-form expression for terminal k, based on the Karush-Kuhn-Tucker conditions and power constraints, is as shown above. It is easy to prove that CMSE is a monotonically increasing function of Lagrange multipliers, and therefore the Lagrange multipliers satisfying the KKT conditions can be obtained using the bisection method.

[0090] Step S2034, the second calculation step, updates the preset noise reduction factor based on the updated signal transmission power:

[0091] In some embodiments, after calculating the optimal power, this disclosure further modifies the preset denoising factor. It is known that CMSE is a convex function of the denoising factor, so the closed expression of the denoising factor can be expressed as the above equation.

[0092] Step S2035: Repeat the first calculation step or the second calculation step at least once until the number of iterations reaches a preset threshold, and determine the updated signal transmission power and the updated preset noise reduction factor as the control input.

[0093] In some embodiments, the optimal power and the aforementioned noise reduction factor are alternately optimized a preset number of times to determine a local optimum, that is, the updated signal transmission power and the updated preset noise reduction factor are determined as the control input.

[0094] In some embodiments, this disclosure also provides a search-based globally optimal solution method. This method obtains the optimal power in the same way as the solution described above. To obtain the globally optimal solution, this method uses a one-dimensional search for denoising factors to find the optimal solution. That is, given different denoising factors, it finds the current minimum CMSE and finally exhaustively compares the results to obtain the globally optimal solution. Specifically, it performs an exhaustive search for each denoising factor, calculates the optimal power and the corresponding CMSE for each denoising factor, and then selects the optimal power and denoising factor corresponding to the minimum CMSE.

[0095] Since the above methods involve searching for a globally optimal solution, the search process is quite cumbersome and time-consuming. Therefore, this disclosure proposes a search-free globally optimal solution.

[0096] Substituting the closed-form expression for the denoising factor into the CMSE solution formula, the optimization model for the above problem becomes equivalent to:

[0097] in,

[0098] Assuming channel |h 1,b |≤|h 2,b |≤...≤|h K,b We then use a case-by-case analysis approach to solve the above problem and optimize the model.

[0099] Without considering transmission rate constraints, the power constraint optimization problem for terminal K with the worst channel conditions can be expressed as: stλ T V1λ=1 further, where I K For the K-order identity matrix and

[0100] The power-constrained optimization problem for terminal K with the worst channel conditions described above is a generalized Rayleigh quotient maximization problem, therefore its closed-form solution can be expressed as:

[0101] Then, starting from terminal k, the transmission power constraint is solved by defining a K*1 vector:

[0102] Equivalent to:

[0103] Where F = diag(I) k-1 ,0), The corresponding optimization problem is expressed as:

[0104]

[0105] in,

[0106] Similarly, this problem is also a generalized Rayleigh quotient maximization problem, and its closed-form solution can be expressed as:

[0107] The above results are iteratively processed until all elements satisfy the power constraint. If the results satisfy the rate constraint, then this solution is the globally optimal solution; otherwise, the focus shifts to solving the rate constraint problem.

[0108] stλ T Hλ=τ.

[0109] This problem is also a generalized Rayleigh quotient maximization problem, and its closed-form solution can be expressed as:

[0110] If the result satisfies the power constraint, then this solution is the global optimal solution. If it does not, then the set of all elements exceeding the power constraint obtained by the above closed-form solution is defined as set O, i.e.:

[0111] Define optimization vector The first element is the sum of all elements in set O, and the rest are the remaining elements of vector λ. The equivalent representation is:

[0112] Where G = diag(0, I K-v ),

[0113] Therefore, the optimization model for the problem with the globally optimal solution searched above is transformed into:

[0114]

[0115] Where U = diag(1,0,…,0),

[0116] make at the same time Further transforming the problem into:

[0117]

[0118] The above problem is a semidefinite programming problem, which is a convex problem. Therefore, CVX can be used to solve it. According to the rank-one decomposition theorem, this problem must have a feasible solution, so the global optimal solution of the original problem is obtained.

[0119] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0120] This disclosure also provides an optimization apparatus for an integrated NOMA communication and in-flight computing system. It should be noted that this optimization apparatus can be used to execute the optimization method for an integrated NOMA communication and in-flight computing system provided in this disclosure. This apparatus is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the apparatus described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0121] The following describes the optimization device for the NOMA communication and over-the-air computing integrated system provided in the embodiments of this disclosure.

[0122] Figure 4 is a structural block diagram of an optimization device for a NOMA integrated communication and over-the-air computing system according to an embodiment of the present disclosure. As shown in Figure 4, the device includes:

[0123] The first construction unit 10 is used to construct an uplink channel model based on the above-mentioned NOMA communication and over-the-air computing integrated system. The uplink channel model is used to simulate the changes in signals during the process of the above-mentioned terminal equipment sending signals to the above-mentioned communication base station and the above-mentioned data fusion center.

[0124] In some embodiments, this disclosure provides an uplink channel model that integrates wireless NOMA (non-orthogonal multiple access) communication and over-the-air computing. The uplink channel model is configured to simulate the changes in signals during the process of the terminal sending signals to the communication base station for NOMA communication and the terminal sending signals to the data fusion center for over-the-air computing.

[0125] The second construction unit 20 is used to construct an optimization problem model based on multiple preset constraints and the above-mentioned uplink channel model. The preset constraints include the minimum mean square error of the above-mentioned uplink channel model for in-flight calculation, the achievable information rate of the uplink channel model for NOMA communication being greater than a preset value, and the signal transmission power of the above-mentioned uplink model being within a preset range.

[0126] In some embodiments, this disclosure sets an optimization objective based on the above-described uplink channel model, namely, minimizing the above-described system air computation error, and constructs a mathematical model for the objective to obtain the above-described optimization problem model.

[0127] The computing unit 30 is used to solve the above-mentioned optimization problem model to obtain the control input of the above-mentioned NOMA communication and airborne computing integrated system, and to control the operation of the above-mentioned NOMA communication and airborne computing integrated system according to the above-mentioned control input. The above-mentioned control input includes at least the above-mentioned signal transmission power and noise reduction factor.

[0128] In some embodiments, by designing the signal encoding of the terminal device and jointly optimizing the transmission power and denoising factor of the terminal device, the above optimization problem model is solved to minimize the mean square error of the system's calculation under the constraints of the maximum transmission power of the terminal device, the system, and the rate, so as to meet the requirements of system computing power and system latency.

[0129] In this embodiment, the first construction unit constructs an uplink channel model based on the aforementioned NOMA communication and over-the-air computing integrated system. The uplink channel model is used to simulate the signal changes during the process of the terminal device sending signals to the communication base station and the data fusion center. The second construction unit constructs an optimization problem model based on multiple preset constraints and the aforementioned uplink channel model. The preset constraints include minimizing the mean square error of the uplink channel model for over-the-air computing, ensuring that the achievable information rate of the uplink channel model for NOMA communication is greater than a preset value, and ensuring that the signal transmission power of the uplink channel model is within a preset range. The calculation unit solves the aforementioned optimization problem model to obtain the control input of the aforementioned NOMA communication and over-the-air computing integrated system, and controls the operation of the aforementioned NOMA communication and over-the-air computing integrated system according to the aforementioned control input. The aforementioned control input includes at least the aforementioned signal transmission power and the noise reduction factor. This disclosure establishes an uplink channel model based on an integrated wireless NOMA communication and over-the-air computing system. Based on this uplink channel model, a data model (i.e., an optimization problem model) is established to minimize the computational error of terminal devices. Optimization is achieved by designing signal encoding for IoT devices to jointly optimize signal transmission power and denoising factors. Under the constraints of maximizing terminal device transmission power and the system's achievable information rate and signal transmission power, the mean square error of the system's computation is minimized. Control parameters for the integrated NOMA communication and over-the-air computing system are determined, and the system's operation is controlled according to these parameters. This method addresses the problem of unreliable signal transmission performance and computational accuracy in the joint deployment of NOMA communication and over-the-air computing.

[0130] To construct the aforementioned uplink channel model, in one optional implementation, the first construction unit includes:

[0131] The first construction module is used to simulate the process of the terminal device encoding the data to be transmitted to obtain the first target signal, and to obtain the first target formula.

[0132] In some embodiments, it is assumed that the integrated wireless NOMA communication and over-the-air computing system deploys a communication base station and a data fusion center, as well as K single-antenna IoT devices simultaneously performing uplink data transmission and over-the-air computing. Based on this, according to the signal encoding of the data to be transmitted by the terminal device during data transmission, the aforementioned first objective formula is constructed to characterize the changes in data during the signal encoding process.

[0133] The second construction module is used to simulate the process of the first target signal being transmitted through the first target channel to obtain the second target signal, and to obtain the second target formula. It also simulates the process of the data fusion center denoising the second target signal and performing over-the-air calculations to obtain the calculation results, and to obtain the third target formula. Finally, it constructs the mean square error calculation formula for the over-the-air calculations performed by the data fusion center, and obtains the fourth target formula. The first target channel is the uplink channel for the terminal device to transmit data to the data fusion center.

[0134] In some embodiments, the second target formula described above is constructed to characterize the signal changes during the transmission of the first target signal by the first target channel. Then, based on the process of the data fusion center using superimposed data from the air interface channel and performing denoising processing on the acquired signal, the third target formula is constructed. Finally, the error calculation formula for the accuracy of in-flight computation is constructed to obtain the fourth target formula.

[0135] The third construction module is used to simulate the process of the first target signal being transmitted through the second target channel to obtain the third target signal, and to obtain the fifth target formula. It also constructs the information reachability rate calculation formula for the communication base station performing NOMA communication and obtains the sixth target formula. The second target channel is the uplink channel for the terminal device to transmit data to the communication base station.

[0136] In some embodiments, the fifth target formula is constructed to characterize the signal changes during the transmission of the first target signal by the second target channel, and then the sixth target formula is obtained based on the calculation formula of the signal reachability rate of each terminal signal obtained by the communication base station using SIC basic decoding for NOMA communication.

[0137] The fourth construction module is used to construct the uplink channel model based on the first objective formula, the second objective formula, the third objective formula, the fourth objective formula, the fifth objective formula, and the sixth objective formula.

[0138] In some embodiments, the uplink channel model is obtained by constructing a channel model based on the first objective formula, the second objective formula, the third objective formula, the fourth objective formula, the fifth objective formula, and the sixth objective formula.

[0139] To construct the aforementioned first target formula, in one optional implementation, the first construction module includes:

[0140] The first construction submodule is used to acquire the data to be transmitted and the first target signal, and to construct the first target formula based on the data to be transmitted and the first target signal.

[0141] Where, x k The first target signal mentioned above, h is the conjugate of the first target channel mentioned above. k,c For the first target channel mentioned above, |h k,c | represents the modulus of the first target channel mentioned above, s k The above is the data to be sent.

[0142] In some embodiments, the data fusion center operates in time-division duplex mode. Before wireless uplink transmission, the data fusion center broadcasts pilot signals to all terminals. Terminal k uses the reciprocity of the wireless channel to estimate the channel h from terminal k to the data fusion center. k,c Terminal k according to channel h k,c The data is encoded to obtain the first target signal mentioned above, and the corresponding formula for the first target is as shown above.

[0143] It is understood that in this disclosure, the single-antenna IoT device sends only one transmission signal each time, which is sent to the communication base station and the data fusion center for signal transmission and over-the-air computing. Therefore, in the process of signal encoding, signal encoding can be performed only according to the data fusion center.

[0144] To construct the aforementioned second, third, and fourth objective formulas, in one optional implementation, the second construction module includes:

[0145] The second construction submodule is used to acquire the first target signal and the second target signal, and construct the second target formula based on the first target signal and the second target signal.

[0146] Among them, y c For the second target signal mentioned above, p k Where n is the signal transmission power, K is the number of terminal devices, and n is the number of terminal devices. c The ambient noise received by the aforementioned data fusion center;

[0147] In some embodiments, the second target formula is constructed based on the influence of the channel on the first target signal during the transmission of the first target signal to the data fusion center via the first target channel.

[0148] The third construction submodule is used to obtain the above calculation results and construct the above third target formula based on the above calculation results and the above second target signal:

[0149] in, Here is an estimate of the above calculation results, and η is the above denoising factor;

[0150] In some embodiments, the data fusion center directly uses the superposition property of the air interface channel to denoise the obtained signal, directly obtaining the calculation results of K terminal devices. The ideal air interface calculation result is: However, in actual calculations, the aforementioned first target signal is unknown to the data fusion center, which only receives the second target signal. Therefore, this disclosure sets up an estimate based on the received second target signal and the denoising factor.

[0151] The fourth submodule is used to construct the mean square error calculation formula for the above data fusion center to obtain the fourth objective formula:

[0152] Where s is the actual value of the above calculation result, and s satisfies When s is unknown, CMSE satisfies in For n c The noise variance.

[0153] In some embodiments, this disclosure constructs a formula for calculating the mean squared error (MSE) of an over-the-air computation in a data fusion center, using the MSE as a criterion for evaluating simulation results. Furthermore, the MSE is defined as shown in the formula above. Since the ideal over-the-air computation result is unknown, this disclosure sets a formula based on... The system's mean square error is calculated. This mean square error is used to assess the performance of the uplink channel model and to correct it.

[0154] To construct the fifth objective formula and the sixth objective formula described above, in one optional implementation, the third construction module includes:

[0155] The fifth construction submodule is used to acquire the first target signal and the third target signal, and construct the fifth target formula based on the first target signal and the third target signal.

[0156] Among them, y b For the third target signal mentioned above, h k,b For the second target channel mentioned above, n b The ambient noise received by the aforementioned communication base stations;

[0157] In some embodiments, the fifth target formula is constructed based on the influence of the channel on the first target signal during the transmission of the first target signal to the communication base station via the second target channel.

[0158] The sixth construction submodule is used to construct the information reachability rate calculation formula for the above-mentioned communication base station to obtain the sixth target formula: R k =log2(1+γ) k );

[0159] Among them, R k For the rate at which the above information can be obtained, γ k The signal-to-dryness ratio of the data received by the aforementioned communication base station, |h k,b | represents the modulus of the second target channel mentioned above. For n b The noise variance, |h k,b |satisfy|h 1,b |≤|h 2,b |≤...≤|h K,b |

[0160] In some embodiments, the communication base station uses walk-through interference cancellation technology to obtain the signal of each terminal for NOMA communication, and assumes that the channel gains are arranged in ascending order, i.e., |h 1,b |≤|h 2,b |≤...≤|h K,b The signal-to-drying ratio of the first K-1 devices satisfies:

[0161] The signal-to-dryness ratio of the Kth device satisfies: Then, the signal-to-dryness ratio of the terminal device k is substituted into the formula: R k =log2(1+γ) k It can calculate the achievable information rate of the k-th device.

[0162] In order to construct the mathematical model, in one optional implementation, the second building unit mentioned above includes:

[0163] The fifth construction module is used to construct the optimization problem model based on the uplink channel model described above:

[0164] Where r is the aforementioned preset value.

[0165] In some embodiments, the CMSE and NOMA communications transmitted by each terminal in the system via pilot air calculations and the rate are both related to the terminal's transmit power p. kThe CMSE is related to the noise reduction factor. Based on the CMSE calculation formula and the achievable information rate calculation formula, an optimization problem model is established, resulting in the model described above. The formulas above represent the problem of minimizing the mean square error of in-flight computation under constraints, system and rate constraints, and transmit power constraints for each terminal, respectively.

[0166] In order to solve the above optimization problem model to obtain the control parameters, in one optional implementation, the above computing unit includes:

[0167] The acquisition module is used to perform the acquisition steps, acquire a preset denoising factor, and transform the above optimization problem model according to the preset denoising factor:

[0168] Where τ=2 r -1;

[0169] In some embodiments, the optimization problem model described above can be solved using a low-complexity local optimum method. Given a denoising factor, i.e., the aforementioned preset denoising factor, the solution to the sub-problem of terminal transmit power can be transformed into the aforementioned model.

[0170] The sixth module is used to execute the construction steps, constructing the Lagrangian function based on the transformed optimization problem model described above:

[0171] Where α is a Lagrange multiplier;

[0172] In some embodiments, the corresponding Lagrangian function, based on the transformed model described above, is represented as shown in the above equation.

[0173] The first calculation module is used to perform the first calculation step, updating the signal transmission power according to the Lagrange function:

[0174] In some embodiments, the optimal power closed-form expression for terminal k, based on the Karush-Kuhn-Tucker conditions and power constraints, is as shown above. It is easy to prove that CMSE is a monotonically increasing function of Lagrange multipliers, and therefore the Lagrange multipliers satisfying the KKT conditions can be obtained using the bisection method.

[0175] The second calculation module is used to perform the second calculation step, updating the preset noise reduction factor based on the updated signal transmission power:

[0176] In some embodiments, after calculating the optimal power, this disclosure further modifies the preset denoising factor. It is known that CMSE is a convex function of the denoising factor, so the closed expression of the denoising factor can be expressed as the above equation.

[0177] The repetition module is used to repeat the first calculation step or the second calculation step at least once until the number of iterations reaches a preset threshold, and to determine the updated signal transmission power and the updated preset noise reduction factor as the control input.

[0178] In some embodiments, the optimal power and the aforementioned noise reduction factor are alternately optimized a preset number of times to determine a local optimum, that is, the updated signal transmission power and the updated preset noise reduction factor are determined as the control input.

[0179] The optimization device for the aforementioned NOMA integrated communication and over-the-air computing system includes a processor and a memory. The first building unit, the second building unit, and the computing unit are all stored as program units in the memory, and the processor executes these program units stored in the memory to achieve the corresponding functions. All of the above modules reside in the same processor; alternatively, the modules may be located in different processors in any combination.

[0180] The processor contains a kernel, which retrieves the corresponding program units from memory. One or more kernels can be configured, and communication accuracy can be improved by adjusting kernel parameters.

[0181] The memory may include non-permanent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0182] This disclosure provides a computer-readable storage medium including a stored program, wherein, when the program is executed, it controls the device containing the computer-readable storage medium to perform the optimization method of the NOMA communication and over-the-air computing integrated system.

[0183] This disclosure provides a processor for running a program, wherein the program executes the optimization method of the NOMA communication and over-the-air computing integrated system.

[0184] This disclosure provides an integrated NOMA communication and in-flight computing system, which includes a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it implements at least the steps of the optimization method for the integrated NOMA communication and in-flight computing system described above.

[0185] This disclosure also provides a computer program product that, when executed on a data processing device, is adapted to perform a program that initializes an optimized method having at least the steps described above for an integrated NOMA communication and over-the-air computing system.

[0186] It is obvious to those skilled in the art that the modules or steps of this disclosure described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. They can be implemented using computer-executable program code, and thus can be stored in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those presented herein, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, this disclosure is not limited to any particular combination of hardware and software.

[0187] Those skilled in the art will understand that embodiments of this disclosure can be provided as methods, systems, or computer program products. Therefore, this disclosure can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this disclosure can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0188] This disclosure is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in one or more flowchart illustrations and / or one or more block diagrams.

[0189] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means that implement the functions specified in one or more flowcharts and / or one or more block diagrams.

[0190] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, such that the instructions, which execute on the computer or other programmable apparatus, provide steps for implementing the functions specified in one or more flowcharts and / or one or more block diagrams.

[0191] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0192] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0193] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic tape, disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0194] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0195] From the above description, it can be seen that the embodiments of this disclosure achieve the following technical effects:

[0196] 1) The optimization method of the NOMA communication and over-the-air computing integrated system disclosed herein comprises: First, constructing an uplink channel model based on the aforementioned NOMA communication and over-the-air computing integrated system. The uplink channel model is used to simulate the signal changes during the process of the terminal device transmitting signals to the communication base station and the data fusion center. Then, constructing an optimization problem model based on multiple preset constraints and the aforementioned uplink channel model. The preset constraints include minimizing the mean square error of the over-the-air computing performed by the aforementioned uplink channel model, achieving an achievable information rate for NOMA communication performed by the aforementioned uplink channel model that is greater than a preset value, and ensuring that the signal transmission power of the aforementioned uplink channel model is within a preset range. Finally, solving the aforementioned optimization problem model yields the control input of the aforementioned NOMA communication and over-the-air computing integrated system, and controlling the operation of the aforementioned NOMA communication and over-the-air computing integrated system based on the aforementioned control input. The aforementioned control input includes at least the aforementioned signal transmission power and a noise reduction factor. This disclosure establishes an uplink channel model based on an integrated wireless NOMA communication and over-the-air computing system. Based on this uplink channel model, a data model (i.e., an optimization problem model) is built to minimize the computational error of the terminal device. Optimization is achieved by designing signal encoding for IoT devices to jointly optimize signal transmission power and denoising factors. Under the constraints of maximizing terminal device transmission power and the system's achievable information rate and signal transmission power, the mean square error of the system's computation is minimized. Control parameters for the integrated NOMA communication and over-the-air computing system are determined, and the system's operation is controlled according to these parameters. This method addresses the lack of a method in the prior art for jointly deploying NOMA communication and over-the-air computing while ensuring signal transmission performance.

[0197] 2) The optimization device for the NOMA communication and over-the-air computing integrated system disclosed herein comprises: a first construction unit constructing an uplink channel model based on the aforementioned NOMA communication and over-the-air computing integrated system; the uplink channel model being used to simulate signal changes during the process of the terminal device transmitting signals to the communication base station and the data fusion center; a second construction unit constructing an optimization problem model based on multiple preset constraints and the aforementioned uplink channel model; the preset constraints including minimizing the mean square error of the uplink channel model for over-the-air computing, the achievable information rate of the uplink channel model for NOMA communication being greater than a preset value, and the signal transmission power of the uplink channel model being within a preset range; and a calculation unit solving the aforementioned optimization problem model to obtain the control input of the aforementioned NOMA communication and over-the-air computing integrated system, and controlling the operation of the aforementioned NOMA communication and over-the-air computing integrated system according to the aforementioned control input; the aforementioned control input including at least the aforementioned signal transmission power and a noise reduction factor. This disclosure establishes an uplink channel model based on an integrated wireless NOMA communication and over-the-air computing system. Based on this uplink channel model, a data model (i.e., an optimization problem model) is built to minimize the computational error of the terminal device. Optimization is achieved by designing signal encoding for IoT devices to jointly optimize signal transmission power and denoising factors. Under the constraints of maximizing terminal device transmission power and the system's achievable information rate and signal transmission power, the mean square error of the system's computation is minimized. Control parameters for the integrated NOMA communication and over-the-air computing system are determined, and the system's operation is controlled according to these parameters. This method addresses the lack of a method in the prior art for jointly deploying NOMA communication and over-the-air computing while ensuring signal transmission performance.

[0198] The above description is merely a preferred embodiment of this disclosure and is not intended to limit this disclosure. Various modifications and variations can be made to this disclosure by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for optimizing NOMA communication and aerial computing integrated system, wherein, The NOMA communication and air computing integrated system comprises a communication base station, a data fusion center and a plurality of terminal devices, comprising: According to the NOMA communication and air computing integrated system, an uplink channel model is constructed, which is used to simulate the change of signals in the process of the terminal device sending signals to the communication base station and the data fusion center; According to a plurality of preset constraint conditions and the uplink channel model, an optimization problem model is constructed, the preset constraint conditions comprising the mean square error and minimum of the uplink channel model for air computing, the achievable information rate of the uplink channel model for NOMA communication being greater than a preset value, and the signal transmission power of the uplink channel model being within a preset range; The optimization problem model is solved to obtain the control input of the NOMA communication and air computing integrated system, and the NOMA communication and air computing integrated system is controlled to operate according to the control input, the control input at least comprising the signal transmission power and a denoising factor.

2. The method of claim 1, wherein, According to the NOMA communication and air computing integrated system, an uplink channel model is constructed, comprising: The process of the terminal device encoding the to-be-sent data to obtain a first target signal is simulated to obtain a first target formula; The process of the first target signal being transmitted through a first target channel to obtain a second target signal is simulated to obtain a second target formula, the process of the data fusion center denoising and performing air computing on the second target signal to obtain a calculation result is simulated to obtain a third target formula, a mean square error calculation formula of the data fusion center performing air computing is constructed to obtain a fourth target formula, and the first target channel is an uplink channel of the terminal device transmitting data to the data fusion center, The process of the first target signal being transmitted through a second target channel to obtain a third target signal is simulated to obtain a fifth target formula, an information achievable rate calculation formula of the communication base station performing NOMA communication is constructed to obtain a sixth target formula, and the second target channel is an uplink channel of the terminal device transmitting data to the communication base station; The uplink channel model is constructed according to the first target formula, the second target formula, the third target formula, the fourth target formula, the fifth target formula and the sixth target formula.

3. The method of claim 2, wherein, The process of the terminal device encoding the to-be-sent data to obtain a first target signal is simulated to obtain a first target formula, comprising: The to-be-sent data and the first target signal are obtained, and the to-be-sent data and the first target The signal builds the first target formula: wherein x k is the first target signal, h is the conjugate of the first target channel k,c h is the first target channel k,c | is the modulus of the first target channel k s is the data to be sent 4. The method of claim 3, wherein, The process of the first target signal being transmitted through a first target channel to obtain a second target signal is simulated to obtain a second target formula, the process of the data fusion center denoising and performing air computing on the second target signal to obtain a calculation result is simulated to obtain a third target formula, a mean square error calculation formula of the data fusion center performing air computing is constructed to obtain a fourth target formula, comprising: acquiring the first target signal and the second target signal, and constructing the second target formula according to the first target signal and the second target signal: wherein y c is the second target signal, p k is the signal transmission power, K is the number of terminal devices, n c is the receiving environmental noise of the data fusion center; obtaining the calculation result, and constructing the third target formula according to the calculation result and the second target signal: wherein The estimated value of the calculation result is η, and η is the denoising factor; The mean square error calculation formula for constructing the data fusion center to perform air computing is obtained as a fourth target formula: where s is the actual value of the calculation result, s satisfies In the case that s is unknown, CMSE satisfies wherein for n c the noise variance.

5. The method of claim 4, wherein, simulate a process of transmitting the first target signal through a second target channel to obtain a third target signal, obtain a fifth target formula, construct a formula for calculating information achievable rate of the communication base station for NOMA communication, obtain a sixth target formula, including: acquiring the first target signal and the third target signal, and constructing the fifth target formula according to the first target signal and the third target signal: wherein y b is the third target signal, h k,b is the second target channel, n b is the receiving environmental noise of the communication base station; The information of the communication base station for constructing the NOMA communication is used to obtain a sixth target formula: R k = log2(1 + γ k ); wherein R k is the information achievable rate, γ k is the signal-to-interference ratio of the communication base station receiving data, |h k,b | is the modulus of the second target channel, for n b the noise variance, |h k,b | satisfies |h 1,b | ≤ |h 2,b | ≤... ≤ |h K,b |.

6. The method of claim 5, wherein, construct an optimization problem model according to a plurality of preset constraint conditions and the uplink channel model, including: constructing the optimization problem model according to the uplink channel model: wherein r is the preset value.

7. The method of claim 6, wherein, solve the optimization problem model to obtain the control input of the NOMA communication and air computing integrated system, including: An acquisition step of acquiring a preset denoising factor, and transforming the optimization problem model according to the preset denoising factor: where τ = 2 r -1; a construction step of constructing a Lagrangian function according to the transformed optimization problem model: wherein a is a Lagrange multiplier; A first calculation step, updating the signal transmission power according to a Lagrangian function: A second calculation step, updating the preset denoising factor according to the updated signal transmission power: repeat the first calculation step or the second calculation step at least once in turn until the iteration number reaches a preset threshold, and determine the updated signal transmission power and the updated preset denoising factor as the control input. 8.An optimization apparatus of a NOMA communication and aerial computing integrated system, wherein, The NOMA communication and air computing integrated system includes a communication base station, a data fusion center and a plurality of terminal devices, and the apparatus includes: A first construction unit is configured to construct an uplink channel model according to the NOMA communication and air computing integrated system, the uplink channel model being used to simulate changes in signals in a process of the terminal device sending signals to the communication base station and the data fusion center. A second construction unit is configured to construct an optimization problem model according to a plurality of preset constraint conditions and the uplink channel model, the preset constraint conditions including a mean square error and a minimum of the uplink channel model for air computing, an achievable information rate of the uplink channel model for NOMA communication being greater than a preset value, and a signal transmission power of the uplink channel model being within a preset range. A calculation unit is configured to solve the optimization problem model to obtain a control input of the NOMA communication and air computing integrated system, and to control the NOMA communication and air computing integrated system to operate according to the control input, the control input at least including the signal transmission power and a denoising factor. The computer readable storage medium includes a stored program, wherein when the program runs, the device where the computer readable storage medium is located executes the method of any one of claims 1-7.

9. A computer readable storage medium, wherein, including: 10.A NOMA communication and aerial computing integrated system, wherein, one or more processors, memories, and one or more programs, wherein the one or more programs are stored in the memories and configured to be executed by the one or more processors, and the one or more programs include programs for executing the method of any one of claims 1-7. ​

Images