A multi-user high-energy-efficiency hybrid resource allocation method suitable for IoV

Abstract

The application discloses a kind of multi-user high energy efficiency hybrid resource allocation method suitable for IoV, for the case that multiple primary users and multiple secondary users exist simultaneously in cognitive vehicle networking, a reasonable and efficient resource allocation scheme is proposed.Each primary user enjoys a separate authorized frequency band, and the secondary user communicates using a hybrid spectrum access method without affecting the quality of service of the primary user, and shares the authorized spectrum with the primary user. Multiple secondary users report the sensing results to the fusion center by local spectrum sensing of all subchannels, while considering the presence of errors in the reporting channel, the false alarm probability and detection probability of the global decision are obtained;Under the constraint conditions of secondary user sending end average transmission power, primary user interference and system rate, the system model of maximum energy efficiency is determined, and the objective function is obtained;Then the objective function is converted into a convex function using distributed programming, the power allocation part uses the Lagrange function, and the optimal solution expression is obtained according to the KKT condition;Subgradient algorithm is used to calculate the Lagrange multiplier, and finally the optimal power allocation of the secondary user on each subchannel is obtained to achieve the purpose of maximizing the energy efficiency of the IoV communication system.

Description

Technical Field

[0001] This invention relates to a cognitive vehicle-to-everything (IoV) technology, and more particularly to a resource allocation method for cognitive vehicle-to-everything (IoV), and more specifically, to a multi-user, high-efficiency hybrid resource allocation method applicable to IoV. Background Technology

[0002] The Internet of Vehicles (IoV) is a network that enables comprehensive connectivity between vehicles and cloud platforms, vehicles themselves, vehicles and roads, vehicles and pedestrians, and within vehicles. With the increasing number of users and the diversification of service applications, while communication brings convenience, it also exacerbates the scarcity of spectrum resources. Especially in scenarios with high traffic density, how to rationally and efficiently utilize spectrum resources has become a significant issue.

[0003] Cognitive Radio Networks (CRNs) address the spectrum shortage problem through intelligent sensing and dynamic spectrum access, integrating cognitive radio technology into vehicle-to-everything (V2X) networks to form cognitive V2X networks, thus making fuller use of spectrum resources. Dynamic spectrum access technologies mainly include Opportunistic Spectrum Access (OSA), Spectrum Sharing (SS), and Hybrid Spectrum Access (HSA). This patent employs hybrid spectrum access technology; that is, from the perspective of the cognitive V2X model, as... Figure 1 As shown, the Primary User (PU) network refers to a traditional wireless communication network consisting of primary users and primary user base stations. It enjoys the highest priority privilege of using licensed frequency bands. The Secondary User (SU) network consists of secondary users, a convergence center, and the primary user base station. It does not have licensed spectrum, but through spectrum sensing technology, it can use licensed spectrum in a hybrid access manner without affecting the primary user's service quality. When the primary user channel is idle, the secondary user uses P... (0) Power transmission data, when the primary user exists, the secondary user transmits data in P... (1) Power transmission data, where P (0) >P (1) .

[0004] Many researchers have considered integrating cognitive radio technology into vehicle-to-everything (V2X) networks. Some literature employs traditional opportunistic spectrum access methods, but these methods have lower spectrum utilization than hybrid approaches. Other literature proposes a cognitive IoV cooperative spectrum sensing power allocation method based on maximizing energy efficiency, but this is only suitable for scenarios with only one primary user and multiple secondary users. Therefore, this invention is applicable to scenarios in cognitive V2X networks where multiple primary and secondary users exist simultaneously, proposing a highly energy-efficient hybrid resource allocation method. Summary of the Invention

[0005] Purpose of the invention: To address the aforementioned problems in existing technologies, this invention proposes a multi-user high-efficiency hybrid resource allocation method suitable for IoV (Internet of Vehicles). This method can achieve reasonable and efficient allocation of secondary user transmission power across multiple primary user channels in cognitive vehicle networks, with the goal of maximizing energy efficiency.

[0006] Technical Solution: Considering the simultaneous presence of multiple primary and secondary users in a cognitive vehicle-to-everything (IoV) network, this solution aims to maximize the energy efficiency of the IoV communication system through rational and efficient resource allocation. Primary users enjoy the highest priority in using licensed spectrum and can use it for communication at any time. Secondary users, without affecting the service quality of primary users, communicate using a hybrid spectrum access method, sharing the licensed spectrum with primary users. N primary users share N licensed sub-channels. Secondary users perform local spectrum sensing on all sub-channels and report the sensing results to the fusion center. The system also considers the possibility of errors in the reported channels, obtaining the false alarm probability and detection probability for global decision-making. Under the constraints of average transmission power at the secondary user transmitter, primary user interference, and system rate, a system model maximizing energy efficiency is determined, yielding the objective function. Distributed programming is then used to transform the objective function into a convex function. For power allocation, a Lagrange function is constructed, and the optimal solution expression is obtained based on the KKT conditions. A sub-gradient algorithm is used to calculate the Lagrange multipliers, finally obtaining the optimal power allocation for secondary users transmitting data on each sub-channel. The above invention is achieved through the following technical solution: a multi-user high-efficiency hybrid resource allocation method suitable for IoV, comprising the following steps:

[0007] (1) Each user performs cooperative spectrum sensing on all sub-channels simultaneously to obtain the false alarm probability and detection probability of global decision. The spectrum sensing error probability is considered in the system model to obtain the expression for the average rate and average total energy consumption.

[0008] (2) Under the constraints of average transmission power at the secondary user transmitter, primary user interference, and system rate, determine the system model that maximizes energy efficiency and obtain the objective function;

[0009] (3) The quasi-concave problem is transformed into a convex function by fractional programming, and a new objective function is obtained. The two transmit powers required for the hybrid spectrum access method are used as variables of the objective function.

[0010] (4) Construct the Lagrange function and obtain the optimal solution according to the KKT conditions;

[0011] (5) The sub-gradient algorithm is used to obtain the optimal Lagrange multiplier, which is then substituted into the optimal solution to complete the optimal power allocation.

[0012] Furthermore, step (1) includes the following specific steps:

[0013] (1a) Assume there are N primary users and M secondary users in the system. The secondary users use energy detection technology to perform local spectrum sensing on N sub-channels and obtain the local false alarm probability of the i-th secondary user on the j-th sub-channel. and local detection probability Where i∈[1,M], j∈[1,N];

[0014] (1b) Considering the case where the reporting channel is not ideal, use P e i Let represent the error probability when the i-th sub-user reports. The fusion center uses the "OR" criterion and assumes that all sub-users transmit in the same but independent reporting channels to obtain the false alarm probability for the global decision of the j-th sub-channel. and detection probability

[0015] (1c) Based on the actual state of the j-th primary user and the result of the spectrum perception of the j-th sub-channel, four combinations are summarized.

[0016] (1d) Based on the probability that the j-th main user does not exist. and the probability of existence It can be calculated The probability in the case and The probability under the given circumstances;

[0017] (1e) yields the transmission rate R of a sub-user within a frame in a sub-channel. G The average rate on the j-th sub-channel is defined as R. j Approximate to R G The transmission rate R of the secondary user in the N sub-channels is obtained. T ;

[0018] (1f) Considering the power consumed by the secondary user's sensing and reporting time slots and the circuit power, the average total energy consumption E of the secondary user within one frame is obtained. T .

[0019] Furthermore, step (3) includes the following specific steps:

[0020] (3a) Using fractional programming, we transform the quasi-concave problem into a convex function and define a new objective function:

[0021]

[0022] Where α is a positive coefficient, the optimization problem is expressed as:

[0023]

[0024] (3b) The optimal solution to expression 22 can be calculated using the following formula:

[0025]

[0026] Dinkelbach's Lemma: If and only if

[0027]

[0028] For a fixed α, the optimal power allocation that maximizes energy efficiency can be found, and then α can be updated until Equation 2 is satisfied, at which point the optimal solution can be obtained.

[0029] Furthermore, step (5) includes the following specific steps:

[0030] (5a) The Lagrange multipliers are updated along the direction of the sub-gradient with a suitable step size δ. In the k-th iteration, the Lagrange multipliers τ, υ, and ε are updated according to the following formulas:

[0031]

[0032] ε k+1 =[ε k -δ k (R T -R min )] + Expression 7

[0033] Where δ k Let [x] be an iteration step size greater than 0. + = max{x, 0}.

[0034] (5b) The optimal power allocation is obtained by iteratively calculating the formula (expression 1-expression 3) until F(α) is less than a certain threshold.

[0035] Beneficial effects: The multi-user high-efficiency hybrid resource allocation method for IoV proposed in this invention takes into account the situation where there are multiple primary users and multiple secondary users in the application scenario. It extends the single channel to multiple channels, makes full use of the spectrum resources of multiple sub-channels, and compares the energy efficiency of the multi-channel secondary user hybrid access method with the traditional opportunistic spectrum access method. It can be seen that the multi-user hybrid resource allocation method for IoV proposed in this invention is superior in maximizing system energy efficiency. Attached Figure Description

[0036] Figure 1 To understand the primary and secondary user network models in the Internet of Vehicles;

[0037] Figure 2 A schematic diagram illustrating the working principle of multiple user perception and multi-channel resource allocation;

[0038] Figure 3 Frame structure for secondary users;

[0039] Figure 4 The graph shows the performance variation of system energy efficiency with reporting channel error under different models;

[0040] Figure 5 The graph shows the performance variation of system energy efficiency with sensing time under different models;

[0041] Figure 6 The graph shows the performance variation of system energy efficiency with the number of users under different models. Detailed Implementation

[0042] The core idea of ​​this invention is to propose a high-efficiency hybrid resource allocation method in cognitive vehicle networking to be applicable to scenarios where multiple primary users and multiple secondary users exist simultaneously.

[0043] The present invention will now be described in further detail.

[0044] Step (1) involves collaborative spectrum sensing for all secondary users to derive the false alarm probability and detection probability for global decision-making. The spectrum sensing error probability is then considered in the system model to obtain expressions for the average rate and average total energy consumption. This includes the following steps:

[0045] (1a) Assume the system has N primary users and M secondary users (e.g., ... Figure 2 N primary users share N licensed sub-channels. When a primary user's channel is idle, a secondary user uses P... (0) Power transmission: When the primary user exists, the secondary user transmits power via P. (1) Power transfer, where P (0) >P (1) The frame structure for secondary users is as follows: Figure 3As shown, where T is the frame length; T s To sense the time, during this time period, M secondary users employ multi-band joint detectors to simultaneously perform local spectrum sensing on N sub-channels using energy detection technology, and each initially determines whether the N primary users exist i∈[1,M], j∈[1,N]; T r The reporting time is defined as the period during which each secondary user completes its local perception report to the fusion center. The fusion center then makes a global decision based on the "OR" fusion criterion; T d The data transmission time is defined as the period during which secondary users access the licensed frequency band to transmit data. Furthermore, it is assumed that all channels in this system model are independent and satisfy flat fading within each frame.

[0046] Since local spectrum sensing uses energy detection technology, it can be deduced that the local false alarm probability of the i-th sub-user for the j-th sub-channel is expressed as:

[0047]

[0048] Meanwhile, the local detection probability of the i-th sub-user for the j-th sub-channel is expressed as:

[0049]

[0050] Where, λ j The energy detection threshold; f represents the noise variance; s γ is the sampling frequency. j The average signal-to-noise ratio for primary users; the Q function is defined as:

[0051]

[0052] (1b) Considering the case where the reporting channel is not ideal, use P e i Let represent the error probability when the i-th secondary user reports. Since the fusion center uses the "OR" criterion and assumes all secondary users transmit in the same but independent reporting channel, the false alarm probability and detection probability of the global decision are expressed as follows:

[0053]

[0054] in, This represents the case where the j-th main user does not exist. This represents the case where the j-th main user exists. This indicates that the fusion center detected that the j-th primary user does not exist. This indicates that the fusion center has detected the existence of the j-th main user.

[0055] (1c) Assume that all secondary users transmit in the same but independent reporting channel and that each secondary user has the same spectrum sensing capability, i.e., P e P i If they are the same, then expressions 11 and 12 can be simplified to:

[0056]

[0057] Based on the actual status of the primary user and the results of spectrum sensing, there are four possible combinations for the sensing results of the j-th sub-channel:

[0058] casel: In other words, both the actual situation and the detection result are that the j-th main user does not exist, and the probability in this case is: At this time, SU is using power P in the j-th sub-channel. (0) Perform data transmission;

[0059] case 2: In other words, the actual situation is that the PU does not exist, but the main user is detected to exist. The probability of this situation is: At this time, SU is using power P in the j-th sub-channel. (1) Perform data transmission;

[0060] case 3: In other words, the actual situation is that the PU exists, but the main user is detected as not existing. The probability of this situation is: At this time, SU is using power P in the j-th sub-channel. (0) Perform data transmission;

[0061] case 4: In other words, both the actual situation and the detection results indicate that the main user exists. The probability of this scenario is expressed as follows: At this time, SU is using power P in the j-th sub-channel. (1) To transmit data.

[0062] in for The probability of the situation, for The probability of the situation, P (0) >P (1) .

[0063] Meanwhile, the send-receive model for the above four cases can be expressed as:

[0064]

[0065] Where k = 1, 2, ..., T dx and y represent the transmitted and received signals, respectively; assuming each h is the channel gain between the secondary user transmitter and the secondary user receiver, and g... ss =|h| 2 n is a noise signal with a mean of 0 and a variance of . s is the fading signal received by the main user, which has a mean of 0 and a variance of .

[0066] (1d) The probability of detecting that the j-th main user does not exist. and the probability of existence It can be calculated that:

[0067]

[0068] Similarly, Probability in the case of as well as Probability in the case of We can calculate the following respectively:

[0069]

[0070] (1e) can be used to obtain the rate R of a single sub-user within a frame. G , represented as:

[0071]

[0072] in Given h and Mutual information between x and y; h e (.) represents the differential entropy; ω is defined as:

[0073]

[0074] The average rate on the j-th sub-channel is represented by R. j R can be defined j For R G An approximation of R, where R is... j It can be represented as:

[0075]

[0076] In the j-th sub-channel, it is assumed that the channel power gain from each secondary user transmitter to the corresponding secondary user receiver is equal, and is denoted as g. ss，j The transmission rate of the secondary user in the N sub-channels is:

[0077]

[0078] (1f), the average total energy consumption of the user in one frame of the j-path subchannel is expressed as:

[0079]

[0080] Where P s and P r These represent the power consumed by the secondary user sensing time slot and the reporting time slot, respectively; P c Circuit power refers to the power consumed by the transmitter circuit (mixer, filter, digital-to-analog converter, etc.).

[0081] Step (2): Under the constraints of average transmission power at the secondary user transmitter, primary user interference, and system rate, determine the system model that maximizes energy efficiency, and obtain the objective function:

[0082] (2a) Energy efficiency is defined as the ratio of the average rate of energy available within a time slot to the average total energy consumption, which can be expressed as:

[0083]

[0084] Expression 15 is our objective function, which optimizes power. and The allocation achieves the maximization of the objective function.

[0085] (2b) Considering the constraints of the objective function, the first is the power constraint. To ensure the power budget for secondary users, the average transmit power of secondary users in the fading channel needs to be constrained to be less than the power threshold, i.e.:

[0086]

[0087] Among them, P max Let P be the average transmission power threshold for the secondary user's transmitter. max value

[0088] (2c) Considering interference constraints, in order to ensure that the primary user's communication is not affected, the interference to the secondary user must be less than the maximum interference threshold that the primary user can tolerate. th , represented as:

[0089]

[0090] Where g sp，j Let I be the channel power gain from the SU transmitter to the PU receiver, and given I th The value of .

[0091] (2d) In order to ensure normal communication for secondary users, their speed must reach a certain level, and this threshold is given as R. min ,Right now:

[0092] R T ≥R min Expression 28

[0093] (2e) Based on the above analysis, the system objective function can be modeled as:

[0094]

[0095] Where C1, C2, and C3 represent the constraints in expressions 26, 27, and 28 above, respectively, and C4 ensures that the power is non-negative.

[0096] Step (3) uses fractional programming to transform the quasi-concave problem into a convex function, obtaining a new objective function. The two transmit powers required for the hybrid spectrum access method are used as variables of the objective function, including the following steps:

[0097] (3a) Using fractional programming, the quasi-concave problem is transformed into a convex function, and a new objective function is defined:

[0098]

[0099] Where α is a positive coefficient, the optimization problem is expressed as:

[0100]

[0101] (3b) The optimal solution to expression 31 can be calculated using the following formula:

[0102]

[0103] Dinkelbach's Lemma: If and only if

[0104]

[0105] When the condition in expression 33 is true, expressions 29 and 31 have the same optimal solution. Therefore, the original optimization problem is transformed into expression 31. For a fixed α, the optimal power allocation that maximizes energy efficiency can be found, and then α is updated until expression 33 is satisfied. At this point, the optimal solution for expression 29 can be obtained.

[0106] Step (4) involves constructing the Lagrangian function and obtaining an approximate optimal solution based on the Karush-Kuhn-Tucker conditions, i.e., the KKT conditions. This includes the following steps:

[0107] (4a), the Lagrange function is expressed as:

[0108]

[0109] Where τ, υ, and ε are the Lagrange multipliers of constraints C1, C2, and C3, respectively, and all three are non-negative. Find the value of Lagrange multipliers for expression 34. and The partial derivative of , and setting the derivative to 0, that is

[0110] (4b) According to the KKT conditions, the optimal transmit power P (0)* P (1)* It can be obtained using the following formula:

[0111]

[0112] Where, [x] + = max{x, 0}.

[0113] Step (5) uses the sub-gradient algorithm to obtain the optimal Lagrange multipliers, substitutes them into the optimal solution, and completes the optimal power allocation, including the following steps:

[0114] (5a) The Lagrange multipliers are updated along the direction of the sub-gradient with a suitable step size δ. In the k-th iteration, the Lagrange multipliers τ, υ, and ε are updated according to the following formulas:

[0115]

[0116] ε k+1 =[ε k -δ k (R T -R min )] + Expression 39

[0117] Where δ k Let [x] be an iteration step size greater than 0. + = max{x, 0}.

[0118] (5b) The optimal power allocation is obtained by iteratively calculating the formula (expression 35-expression 39) until F(α)≤θ1 holds.

[0119] Finally, simulations were performed, and the results were analyzed and compared. Figure 4 The simulation graphs of the energy efficiency of the traditional opportunistic and the multi-user hybrid cooperative spectrum sensing models proposed in this invention as a function of the reporting channel error probability were compared and analyzed. It is clear from the graphs that the system's energy efficiency increases with P... e The system energy efficiency decreases as the probability of a reported channel error increases, meaning that the higher the probability of a reported channel error, the lower the system energy efficiency, thus proving the impact of the reported channel error probability on system energy efficiency; meanwhile, with P fixed... e It can be seen that the hybrid energy efficiency we proposed is superior to the traditional opportunistic energy efficiency.

[0120] Figure 5A comparative analysis was conducted on the energy efficiency of traditional opportunistic and hybrid collaborative spectrum sensing models as a function of sensing time T. s The simulation diagram shows the changes. The diagram shows the same model and the same temperature (T). s Under these conditions, the secondary user rate R min The larger the value of R, the lower the energy efficiency. This is because secondary users need to consume more energy to meet the high speed, thus causing energy efficiency to decrease with increasing speed R. min The energy efficiency decreases as the energy consumption increases. Similarly, under the same conditions, there is a certain gap between hybrid and opportunistic energy consumption, and the former achieves greater energy efficiency than the latter, proving that hybrid energy consumption outperforms opportunistic energy consumption in terms of energy efficiency.

[0121] Figure 6 Simulation plots comparing the energy efficiency of traditional opportunistic and hybrid cooperative spectrum sensing models as a function of the number of secondary users were analyzed. The plots show that the system energy efficiency initially increases and then decreases with the increase in the number of secondary users, with an optimal number of secondary users K=3. This is because as the number of secondary users increases, the improvement in spectrum sensing performance outweighs the performance loss caused by system transmission time and energy consumption. However, as the number of secondary users further increases, energy consumption increases, but spectrum sensing performance tends to stabilize. Therefore, the performance loss caused by energy consumption outweighs the improvement in spectrum sensing performance, resulting in a downward trend in the overall system energy efficiency. Meanwhile, and Figure 4 , Figure 5 Similarly, P e At the same time, we can see that hybrid spectrum access is more energy efficient than traditional opportunistic access.

[0122] Based on the description of the present invention, those skilled in the art should readily recognize that the efficient joint resource allocation algorithm of the present invention can greatly reduce algorithm complexity and ensure system performance.

[0123] The contents not described in detail in this application are existing technologies known to those skilled in the art.

Claims

1. A multi-user, high-efficiency hybrid resource allocation method for IoV (Internet of Vehicles) in cognitive vehicle networking, characterized in that: The steps include the following: (1) Construct a system model containing N primary users and M secondary users. Primary users have unlimited access to licensed frequency bands, while secondary users use licensed frequency bands in a hybrid manner through spectrum sensing. When a primary user's channel is detected to be idle, the secondary user uses P... (0) Power transmission: when the presence of the primary user is detected, the secondary user transmits power at P... (1) Power transfer, where P (0) >P (1) Each secondary user simultaneously performs cooperative spectrum sensing on all sub-channels. The local spectrum sensing of the secondary user adopts energy detection technology to obtain the local false alarm probability of the i-th secondary user for the j-th sub-channel. and local detection probability Considering the case where the reporting channel is not ideal, let the error probability of the i-th secondary user's report be... The fusion center uses the "OR" criterion for global decision-making and assumes that all sub-users transmit in the same but independent reporting channels to obtain the false alarm probability of the global decision for the j-th sub-channel. and detection probability Based on the actual state of the j-th primary user and the results of spectrum sensing of the j-th sub-channel, four possible sensing state combinations are summarized. Based on these four state combinations—"both the actual situation and the detection result are that the primary user does not exist," "the actual situation is that the primary user does not exist but the primary user is detected to exist," "the actual situation is that the primary user exists but the primary user is not detected to exist," and "both the actual situation and the detection result are that the primary user exists"—the probability of detecting that the primary user does not exist in the j-th sub-channel is calculated. and the probability of existence Further, the probabilities of the two scenarios—"the main user exists but the detection shows that the main user does not exist" and "both the actual situation and the detection result indicate that the main user exists"—are derived. and The probability of the following situation; based on the actual occurrence probability of secondary user transmission, calculate the transmission rate R of the secondary user in one frame within a sub-channel. G The average rate on the j-th sub-channel is defined as R. j Approximate to R G The transmission rate R of the secondary user in the N sub-channels is obtained. T Considering the power consumed by the secondary user's sensing and reporting time slots, as well as the circuit power, the average total energy consumption E of the secondary user within one frame is obtained. T ; (2) Under the constraints of average transmission power at the secondary user transmitter, primary user interference, and system rate, determine the system model that maximizes energy efficiency and obtain the objective function; (3) The quasi-concave problem is transformed into a convex function by fractional programming, and a new objective function is obtained. The two transmit powers required for the hybrid spectrum access method are used as variables of the objective function. (4) Construct the Lagrange function and obtain the optimal solution according to the KKT conditions; (5) The sub-gradient algorithm is used to obtain the optimal Lagrange multiplier, which is then substituted into the optimal solution to complete the optimal power allocation.

Images