A method and system for optimizing network service reliability performance of a general sense integration

Abstract

The present application relates to a kind of integrated network service reliability performance optimization methods of sensing and computing, comprising the following steps: step S1. the expression of communication, sensing, computing error probability is obtained and weight coefficient is obtained;Step S2. constraint condition is set;Step S3. the optimization model with the minimization of overall error probability as target is constructed;Step S4. the optimization model is solved, and the optimal solution of integrated network service reliability of sensing and computing is obtained.The present application can effectively improve the reliability of integrated network of sensing and computing, and the overall error rate of frame is used to weigh the performance indicators of communication sensing and computing simultaneously, the optimization method is expected to reveal the reachable performance of sensing and computing in the studied scene, and help the development and application of integrated sensing and computing technology.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of communication, in particular to a sensing and communication integrated network service reliability performance optimization method and system. BACKGROUND

[0002] With the maturity and large-scale commercialization of 5G technology, the research and development of 6G technology has gradually started worldwide. 6G technology has high reliability (99.99999%), large-scale ultra-low latency (10-100us), and high scalability, which can meet the demand for ultra-low latency and high reliability. Among them, sensing and communication integration is considered one of the key technologies of 6G. In the sensing and communication integrated network, sensing and communication share the same hardware and spectrum resources. Compared with pure communication or pure sensing network, sensing and communication integrated network has two important advantages: 1) integration gain, which can effectively utilize limited resources to realize the dual functions of communication and sensing, 2) cooperation gain, which balances the performance of the two functions and makes them mutually assist each other. Thanks to these two advantages, the application research of sensing and communication integration has expanded to many emerging fields, including vehicle networking, environmental monitoring, Internet of Things, and human activity recognition.

[0003] However, there are still many problems to be solved in sensing and communication integration. First, there is a lack of unified integrated performance indicators, and second, the performance boundary under the integrated performance indicators has not been clearly defined. One purpose of studying the integrated performance boundary is to analyze the trade-off relationship between communication and sensing performance. This is because in integrated signal design, communication and sensing have a resource competition relationship. Trade-off analysis can help us make a reasonable choice between non-orthogonal multiplexing signals and orthogonal multiplexing signals. Therefore, it is particularly important to establish a unified integrated performance indicator that meets high reliability and low latency, and to analyze the performance boundary under the integrated performance indicator, which not only reveals the achievable performance of sensing and communication integration in the studied scenario, but also helps the development and application of sensing and communication integration technology.

[0004] Based on this, we propose a sensing and communication integrated network service reliability performance optimization method. First, the expressions of communication, sensing, and computing error probabilities are obtained, and the weight coefficients of power and time are proposed. Then the constraint conditions are set. Finally, an optimization model is constructed to minimize the overall error probability, and the optimization model is solved to obtain the optimal solution of the sensing and communication integrated network service reliability. SUMMARY

[0005] The present application proposes a sensing and communication integrated network service reliability performance optimization method for the scenario of 6G sensing and communication integrated edge computing. This method not only reveals the achievable performance of sensing and communication integration in the studied scenario, but also helps the development and application of sensing and communication integration technology. It has the characteristics of low latency, high reliability, low energy consumption, and high precision.

[0006] To address the aforementioned technical problems, the present invention adopts the following technical solution: a method for optimizing the reliability and performance of integrated communication, sensing, and computing network services, comprising the following steps:

[0007] Step S1. Obtain the expressions for communication, sensing, and error probability calculation, and obtain the weight coefficients;

[0008] Step S2. Set constraints;

[0009] Step S3. Construct an optimization model with the objective of minimizing the overall error probability;

[0010] Step S4. Solve the optimization model to obtain the optimal solution for the reliability of the integrated communication and computing network services.

[0011] Furthermore, step S1 specifically includes the following sub-steps:

[0012] Step S11. Construct a frame structure integrating synesthesia and computation;

[0013] Step S12. Based on the frame structure, calculate the communication, sensing, and error probability expressions respectively;

[0014] Step S13. Based on the obtained communication, sensing, and calculation error probability expressions, set the weighting coefficients for power and time;

[0015] Step S14. Calculate the power and time of the next frame of communication based on the obtained power and time weighting coefficients.

[0016] Furthermore, in step S11, the specific method for constructing the integrated synesthetic and computational frame structure is as follows:

[0017] Define the computation time of the integrated frame structure as t. comp The communication time is t. comm The perception time is divided into two parts: perception + calculation time, which is t. sens The sensing and communication time is t. s+c , where t comp =t sens , t comm =t s+c The total time of a frame is T. Since only the communication and sensing processes require signal transmission and power consumption, the power is divided into two parts: sensing + calculation time t. sens The corresponding pure sensing power is p sens Sensing + Communication Time t s+c The power consumed during the corresponding communication and sensing multiplexing process is p. s+c .

[0018] Further, in the step S12, according to the coding error rate of the limited code length, the expression of the communication error probability is:

[0019]

[0020] wherein, ∈ comm is the communication coding error probability, d is the bit number of the data packet transmitted in the wireless communication system using the limited code length coding, is the Q function, V = 1-(1+γ) -2 is the fixed gain of the channel, is the Shannon capacity, wherein, γ is the signal-to-noise ratio, h is the channel gain, σ 2 is the channel noise power;

[0021] According to the extreme value theory, the expression of the calculation error probability is:

[0022] ∈ comp = (1-F D (t th ))(1-G(t comp ;σ,ξ)) (2)

[0023] wherein, ∈ comp is the calculation error probability, the calculation error probability uses the extreme value theory to constrain the queue length, t th is the threshold of the calculation time, F D (t th ) is the cumulative distribution function, σ is the scale parameter, ξ is the shape parameter, G(t comp ;σ,ξ) is the generalized Pareto distribution;

[0024] According to the outage probability, the expression of the perception error probability is:

[0025]

[0026] wherein, ∈ sens is the perception error probability, τ max is the maximum delay, P is the outage probability.

[0027] Further, in the step S13, the communication duration weight coefficient is:

[0028]

[0029] The communication power weight coefficient is:

[0030]

[0031] wherein, w commt is the current frame communication duration weight coefficient, λ is the allocation coefficient, This represents the communication error probability of the previous frame. This represents the perceptual error probability of the previous frame. ε is the error probability calculated in the previous frame. max The maximum error probability in communication, sensing, and computation. T is the communication time of the previous frame. max The maximum delay of a frame. E represents the communication power of the previous frame. max The maximum energy of a frame.

[0032] Furthermore, in step S14:

[0033] Based on the obtained power and time weighting coefficients, the time and power of the current frame communication are calculated respectively.

[0034]

[0035] Furthermore, in step S2:

[0036] The maximum total user latency is T. max Then we have t comm +t comp ≤T max The maximum total energy consumption for users is E. max Then we have t sens p sens +t s+c p s+c ≤E max The probability of errors in communication, sensing, and computation is at most ∈ max Then we have ∈ comm ≤∈ max ,∈ comp ≤∈ max ,∈ sens ≤∈ max ,in.

[0037] Furthermore, in step S3:

[0038] Under constraints of communication error rate, perception error rate, computation error rate, total user latency, and total energy consumption, with the optimization objective of minimizing the overall error rate of a frame, the reliability optimization model for integrated communication, sensing, and computing network services is as follows:

[0039]

[0040] t comm +t comp ≤T max (6b)

[0041] t sens p sens +ts+c p s+c ≤E max (6c)

[0042] ∈ comm ≤∈ max (6d)

[0043] ∈ comp ≤∈ max (6e)

[0044] ∈ sens ≤∈ max (6f)

[0045] Further, in the step S4: according to the constructed optimization model with the minimum of the overall error probability as the target, the matlab is used for solving, and the optimal solution of the sensing and computing integrated network service reliability is obtained.

[0046] The application also provides a sensing and computing integrated network service reliability performance optimization system, comprising:

[0047] An expression acquisition unit is configured to acquire a communication, sensing and computing error probability expression.

[0048] An optimization problem acquisition unit is configured to construct an optimization model with the minimum of the overall error probability as the target.

[0049] An optimal allocation acquisition unit is configured to solve the optimization model and obtain the optimal sensing and computing integrated network service reliability.

[0050] Compared with the prior art, the application has the following beneficial effects:

[0051] The sensing and computing integrated network service reliability optimization model constructed by the application is complete, and the error probabilities of the communication, sensing and computing three parts are comprehensively considered, the minimum of the overall error probability is taken as the optimization target, and the reliability of the sensing and computing integrated network service is further effectively improved.

[0052] The weight coefficients of power and time are introduced, the power and time allocation results of the last frame are affected to the allocation of the next frame, the algorithm is more stable and better, the optimal power and time allocation strategy is obtained, and the application value is high. BRIEF DESCRIPTION OF DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the required drawings in the embodiments or prior art description are briefly introduced below. Obviously, the following drawings and their descriptions merely constitute some embodiments of the present application, and for those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0054] Figure 1 A schematic diagram of a sensing-computing integrated network service architecture is shown in the figure.

[0055] Figure 2 A schematic diagram of a sensing-computing integrated frame structure is shown in the figure. DETAILED DESCRIPTION

[0056] The present application is further described below in conjunction with the drawings: In order to facilitate those skilled in the art to understand and implement the present application, the present application is further described in detail below in conjunction with the embodiments. It should be understood that the embodiments described herein are merely used to illustrate and explain the present application, and are not used to limit the present application.

[0057] For the reliability requirement of sensing-computing integration in the edge computing scenario, the time weight coefficient and the power weight coefficient are introduced on the basis of considering the sensing-computing resource, time delay, energy consumption, error rate and other constraint limitations. The present application aims to disclose the achievable performance of sensing-computing integration in the studied scenario, and to help the development and application of sensing-computing integration technology. It has the characteristics of low time delay, high reliability, low energy consumption and high precision.

[0058] Step 1: Obtain the expression of the communication, sensing and computing error probability and obtain the weight coefficient;

[0059] Step 1: Obtain the expression of the communication, sensing and computing error probability and obtain the weight coefficient, which is as follows:

[0060] (I) Construct the frame structure of sensing-computing integration:

[0061] Figure Two A schematic diagram of the sensing-computing integrated frame structure in the present application is shown in the figure. A frame structure is defined as shown in the figure. Figure Two Among them, the sensing of the current frame is for the next frame, and the computing and communication use the information sensed by the last frame. Among them, the computing time is t comp , the communication time is t comm , the sensing time is divided into two parts, the sensing+computing time t sens , and the sensing+communication time t s+c , wherein t comp =t sens , tcomm = t s+c The total time of one frame is T. Since only the communication and sensing processes need to transmit signals to consume power, the power is divided into two parts, the sensing + computing time t sens The corresponding pure sensing power p sens The sensing + communication time t s+c The corresponding sensing + communication power p s+c .

[0062] (ii) Obtain the expression of the communication, sensing, and computing error probability:

[0063] According to the coding error rate of the limited code length, the expression of the communication error probability is obtained as follows:

[0064]

[0065] where ∈ comm is the communication coding error probability, t comm is the time used for communication in one frame, p s+c is the power consumed in the process of communication and sensing multiplexing, d is the number of bits of the transmitted data packet in the wireless communication system using the limited code length coding, is the Q function, V = 1-(1+γ) -1 is the fixed gain of the channel, is the Shannon capacity, where γ is the signal-to-noise ratio, h is the channel gain, σ 2 is the channel noise power;

[0066] According to the extreme value theory, the expression of the computing error probability is obtained as follows:

[0067] ∈ comp = (1-F D (t th ))(1-G(t comp ; σ, ξ)) (2)

[0068] where ∈ comp is the computing error probability, the queue length is constrained by the extreme value theory, t th is the threshold value of the computing time, F D (t th ) is the cumulative distribution function, σ is the scale parameter, ξ is the shape parameter, and G(t comp ; σ, ξ) is the generalized Pareto distribution.

[0069] According to the outage probability, the expression of the sensing error probability is obtained as follows:

[0070]

[0071] where ∈ sens is the sensing error probability, t sens is the sensing + computing time, t s+c is the sensing + communication time, p sens is the pure sensing power, p s+c is the sensing + communication power, τ max is the delay maximum, P is the outage probability.

[0072] (Three) According to the obtained communication, sensing, and computing error probability expressions, set the weight coefficients of power and time:

[0073] Since the sensing of the last frame is for the calculation and communication of the current frame, the influence of the power and time allocation of the last frame on the current frame is considered, and the power and time weight coefficients are introduced. Among them, the communication time weight coefficient of the current frame includes an error probability factor and an allocation factor. The error probability factor is positively correlated with the communication error probability of the last frame and negatively correlated with the sensing and computing error probabilities, that is, the larger the communication error probability of the last frame, the longer the communication time of the current frame should be, so as to ensure that the communication error probability of the next frame is relatively small, and divided by ε max for normalization; the allocation factor is negatively correlated with the total time of the communication of the last frame in the entire frame, and divided by T max for normalization; therefore, the communication time weight coefficient can be obtained:

[0074]

[0075] The communication power weight coefficient of the current frame includes an error probability factor and an allocation factor. The error probability factor is positively correlated with the communication error probability of the last frame and negatively correlated with the sensing and computing error probabilities, that is, the larger the communication error probability of the last frame, the larger the communication power of the current frame should be, so as to ensure that the communication error probability of the current frame is relatively small, and divided by ε max for normalization; the allocation factor is negatively correlated with the total energy of the communication of the last frame in the entire frame, and divided by E max for normalization; therefore, the communication power weight coefficient can be obtained:

[0076]

[0077] (Four) According to the obtained weight coefficients of power and time, calculate the power and time of the communication of the current frame:

[0078] According to the obtained weight coefficients of power and time, calculate the time and power of the communication of the current frame, respectively, that is,

[0079] Step 2, set the constraint conditions;

[0080] The total delay of the user is maximized as T max , then t comm +tcomp ≤T max ; the maximum total energy consumption of the user is E max , then t sens p sens +t s+c p s+c ≤E max ; the maximum communication, sensing, and computing error probability is ∈ max , then ∈ comm ≤∈ max , ∈ comp ≤∈ max , ∈ sens ≤∈ max .

[0081] Step 3: Construct an optimization model with the minimum overall error probability as the objective;

[0082] Under the constraints of communication error rate, sensing error rate, computing error rate, total user delay, and total energy consumption, the optimization objective is to minimize the overall error rate of one frame, and the integrated network service reliability optimization model is:

[0083]

[0084] t comm +t comp ≤T max (6b)

[0085] t sens p sens +t s+c p s+c ≤E max (6c)

[0086] ∈ comm ≤∈ max (6d)

[0087] ∈ comp ≤∈ max (6e)

[0088] ∈ sens ≤∈ max (6f)

[0089] Wherein, (6a) is the optimization objective, by adjusting the allocation of time and power within one frame, the overall error probability is minimized; (6b) is the time constraint, the maximum time of one frame is T max ; (6c) is the energy constraint, the maximum energy consumed by one frame is E max ; (6d), (6e), and (6f) are respectively the communication, computing, and sensing error probability constraints, and the maximum error probability is ∈ max .

[0090] The original problem (OP) is a non-convex problem, which is difficult to solve effectively. In order to solve this problem, we provide the following theorem.

[0091] Theorem 1. There exists an optimal solution of OP Then holds.

[0092] Proof: We can prove this theorem by contradiction. We assume that there exists only optimal solution t' comm , t' comp , then t' comm + t' comp < T max . Because this solution is optimal, then max t，p (1 - ∈ comm (t' comm ))(1 - ∈ comp (t' comp ))(1 - ∈ sens ) is the global maximum, i.e. Let t" comm = t' comm , t" comp = T max - t' comp , then we can get: t" comp > t' comp , by (1) (6), ∈ comn (t" comm ) = ∈ comn (t' comm ), ∈ comp (t" comp ) < ∈ comp (t' comp ), then violates the assumption of the optimal solution, so there exists an optimal solution of OP Then holds.

[0093] Theorem 2. Given a There exists an optimal solution of OP Then holds.

[0094] Proof: We can prove this theorem by contradiction. We assume that given a t' s , t' s+c , there exists only optimal solution p' sens , p' s+c , then t' sens p' senst' s+c p' s+c < E max Since this solution is optimal, max t，p (1 - ∈ comm (p' sens ))(1 - ∈ comp )(1 - ∈ sens (p' s+c )) is the global maximum, i.e., max t，p (1 - ∈ comm (p' s+c ))(1 - ∈ comp )(1 - ∈ sens (p' s+c , p' sens )) ≥ max t，p (1 - ∈ comm (p" s+c ))(1 - ∈ comp )(1 - ∈ sens (p" s+c , p" sens )), let p" s+c = p' s+c , p" sens = (E max - t' s+c p' s+c ) / t' sens , then we have p" sens > p' sens , and from (2) (6), ∈ comm (p" s+c ) = ∈ comm (p' s+c ), ∈ sens (p" s+c , p" sens ) < ∈ sens (p' s+c , p' sens ), so max t，p (1 - ∈ comm (p' s+c ))(1 - ∈ comp )(1 - ∈ sens (p' s+c , p' sens) ) < max t，p (1 - ∈ comm (p" s+c ))(1 - ∈ comp )(1 - ∈ sens (p" s+c , p" sens )) violates the assumption of the optimal solution. So given a There is an optimal solution of OP Then Holds.

[0095] To solve the above problems, the original problem is degraded into sub-problems SP by applying the above two theorems.

[0096]

[0097] t comm +t comp =T max (7b)

[0098] t sens p sens +t s+c p s+c =E max (7c)

[0099] ∈ comm ≤∈ max (7d)

[0100] ∈ comp ≤∈ max (7e)

[0101] ∈ sens ≤∈ max (7f)

[0102] Step 4: Solve the optimization model to obtain the optimal solution of the integrated network service reliability.

[0103] 1. According to the constructed optimization model with the minimum overall error probability as the target, global search is used to solve the integrated network service reliability optimal solution. The specific solving process is as follows: first, set the upper and lower bounds of the search to determine the search range. Specifically, set T max = 25ms, E max = 800mJ, the search range of t comm is [0, 25ms], t comp =T max -t comm , the search range of p s+c is [0, 32dbm], p sens =(E max -t comm p s+c ) / t sens , the initial value of the communication weight is w commt =1, w commp =1, and the frame number is k=1.

[0104] 2. If k=1, For the initial value of our search, go to 3; if k > 1, then calculate the time weight coefficient w of communication according to formula (4) (5) commt And the power weight coefficient w of communication commp , For the initial value of our search.

[0105] 3. Then, according to the objective function (7a), the objective function should return the error probability under given t and p, then, call the err function to calculate the error probability and return it.

[0106] 4. Finally, use the fminsearch function to perform global search, and return the value of t comm , t comp , p s+c , p sens Corresponding to the minimum error probability, that is, our optimal solution.

[0107] 5. Record the error probability ∈ comm , ∈ comp , ∈ sens Corresponding to the optimal solution for the weight coefficient calculation of the next frame.

[0108] The embodiment also provides a sensing and computing integrated network service reliability performance optimization system, which comprises:

[0109] An expression acquisition unit is configured to acquire a communication, sensing and computing error probability expression.

[0110] An optimization problem acquisition unit is configured to construct an optimization model with the minimization of overall error probability as a target.

[0111] An optimal allocation acquisition unit is configured to solve the optimization model to obtain a sensing and computing integrated network service reliability optimal solution.

[0112] It should be understood that the above description of the preferred embodiments is more detailed, and therefore should not be considered as a limitation on the scope of patent protection of the present application. Those skilled in the art can make substitutions or modifications without departing from the scope of the claims of the present application, and all fall within the scope of protection of the present application. The scope of protection of the present application should be subject to the appended claims.

Claims

1. A method for optimizing network service reliability performance of a network integrated with a sensing algorithm, characterized in that, Comprising the following steps: Step S1. Obtain the expression of communication, perception, and calculation error probability and obtain the weight coefficient; comprising the following sub-steps: Step S11. Build a frame structure integrated with communication, perception, and calculation; the specific method is as follows: The computation time of the integrated frame structure is defined as , the communication time is , the sensing time is divided into two parts, the sensing + computation time is , and the sensing + communication time is , wherein , , the total time of a frame is T, since only the communication and sensing processes need to transmit signals to consume power, the power is divided into two parts, the sensing + computation time , the corresponding pure sensing power is , the sensing + communication time , and the power consumed in the corresponding communication and sensing multiplexing process is ; Step S12. According to the frame structure, calculate the communication, perception, and calculation error probability expression respectively; Step S13. According to the obtained communication, perception, and calculation error probability expression, set the weight coefficient of power and time; Communication duration weight coefficient: Communication power weight coefficient: wherein is a current frame communication duration weight coefficient, is an allocation coefficient, is a communication error probability of a previous frame, is a perception error probability of a previous frame, is a calculation error probability of a previous frame, is a maximum error probability of communication, perception, calculation, is a communication time of a previous frame, is a maximum time delay of a frame, is a communication power of a previous frame, is a maximum energy of a frame; Step S14. According to the obtained weight coefficient of power and time, calculate the power and time of the current frame communication; Step S2. Set the constraint condition; Step S3. Build an optimization model with the minimum overall error probability as the target; Step S4. Solve the optimization model to obtain the optimal solution of the integrated communication, perception, and calculation network service reliability.

2. The method according to claim 1, wherein In step S12, according to the coding error rate of limited code length, the communication error probability expression can be obtained as: wherein, is a communication coding error probability, d is a number of bits of a data packet transmitted in a wireless communication system employing a finite code length coding, is a Q function, is a fixed gain of a channel, is a Shannon capacity, wherein, , is a signal-to-noise ratio, is a channel gain, is a channel noise power; According to the extreme value theory, the calculation error probability expression can be obtained as: wherein, to calculate the error probability, the error probability is calculated using the extreme value theory to constrain the queue length, to calculate the threshold of the time, is the cumulative distribution function, is the scale parameter, is the shape parameter, is the generalized Pareto distribution; According to the outage probability, the perception error probability expression can be obtained as: wherein, is the perceived error probability, is the maximum delay, P is the outage probability.

3. The method of claim 1, wherein the method further comprises: In step S14: According to the obtained weight coefficient of power and time, calculate the power and time of the current frame communication: According to the obtained power and time weight coefficients, the time and power of the current frame communication are calculated respectively, that is , .

4. The method of claim 1, wherein the method further comprises: In step S2: The maximum total latency of the user is Then, we have The maximum total energy consumption of the user is Then, we have ; The communication, perception, calculation error probability is maximum for Then , , .

5. The method according to claim 4, wherein, In step S3: Under the constraints of communication error rate, perception error rate, calculation error rate, total user delay, and total energy consumption, the optimization target is to minimize the overall error rate of a frame, and the integrated communication, perception, and calculation network service reliability optimization model is:

6. The method according to claim 4, wherein, In step S4: According to the optimization model built with the minimum overall error probability as the target, solve it with matlab to obtain the optimal solution of the integrated communication, perception, and calculation network service reliability.

7. A network service reliability performance optimization system integrating communication, sensing, and computing, characterized in that, Comprise: Expression acquisition unit, used for obtaining the expression of communication, perception, and calculation error probability; Optimization problem acquisition unit, used for building an optimization model with the minimum overall error probability as the target; Optimal allocation acquisition unit, used for solving the optimization model to obtain the optimal solution of the integrated communication, perception, and calculation network service reliability; The integrated communication, perception, and calculation network service reliability performance optimization system is used to execute the steps in the integrated communication, perception, and calculation network service reliability performance optimization method of any one of claims 1-6.

Images