Resource allocation method in case of coexistence of multiple wireless body area networks

Abstract

The present application relates to the technical field of wireless communication, and particularly relates to a resource allocation method under a multi-wireless body area network coexistence condition, comprising the following steps: constructing a multi-wireless body area network system model based on bidirectional transmission of information and energy, wherein the multi-wireless body area network system model comprises a pre-arranged radio frequency energy source for wireless energy supply, a plurality of freely movable wireless body area networks and an AP equipped with an edge server; constructing an information and energy transmission protocol within the wireless body area network system model under the multi-wireless body area network coexistence condition; constructing an overall optimization model based on a transmission protocol selection vector a and a sensor transmission power allocation ratio parameter ρ ij and a time slot sequence P; and solving the optimal configuration of the multi-wireless body area network system model based on the overall optimization model to obtain an optimal transmission strategy. The present application guarantees high reliability of information transmission and energy transmission, especially high reliable transmission of important information.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wireless communication, and more particularly to a resource allocation method in a multi-wireless body area network coexistence scenario. BACKGROUND

[0002] With the vigorous development of wireless communication and Internet of Things technology, and the increasing aging of the population, wireless body area network (WBAN, Wireless Body Area Network) as a sensor technology that can monitor human life and health with high reliability, flexibility, scalability and low cost has been widely concerned and applied. The application of wireless body area network can effectively alleviate the contradiction between limited medical resources and the growing demand of the aging population, and has been widely used in electronic health care, providing real-time and continuous care for health monitoring.

[0003] Wireless body area networks can be widely deployed in densely populated scenarios, such as hospital wards, waiting rooms, nursing homes, or smart homes. However, when the communication areas of adjacent wireless body area networks are close to or even overlap with each other, the system will inevitably suffer from serious intra-network or inter-network interference. Intra-body area network interference refers to the interference caused by simultaneous data transmission of nodes within the network; inter-network interference is caused by simultaneous data transmission between two or more adjacent networks. Inter-body area network interference reduces the reliability and timeliness of physiological data transmission, and increases the economic cost of network management and health care. At the same time, simultaneous interference may cause incomplete or outdated emergency medical diagnosis data, thereby threatening people's life safety. When there is interference between networks, it is necessary to take effective interference mitigation solutions, but there are few solutions in existing research to mitigate interference between body area networks.

[0004] At the same time, in most cases, in order to meet the requirements of miniaturization and light weight, the battery size of the sensor is therefore limited, and the energy storage capacity is small. Especially for implanted sensor nodes in the human body, it is very inconvenient to charge or replace them, so long-term continuous and stable power supply of wireless sensor networks has become a key problem. Wireless power transmission (WPT, Wireless Power Transmission) has become an ideal solution to solve the power supply problem of wireless sensors. The proposal of wireless power communication network (WPCN, Wireless Powered Communication Network) realizes the good integration of sensor network and WPT technology, making it possible to use wireless power sensors to collect data in wireless sensor networks.

[0005] How to ensure the high reliability of bidirectional transmission of information and energy in the multi-network coexistence scenario has become a technical problem that needs to be solved by those skilled in the art. SUMMARY

[0006] Therefore, the application provides a resource allocation method in a multi-wireless body area network coexistence situation, which guarantees high reliability of information transmission and energy transmission, especially high reliable transmission of important information.

[0007] To achieve the above-mentioned purpose, the application adopts the following technical solutions:

[0008] A resource allocation method in a multi-wireless body area network coexistence situation, comprising the following steps:

[0009] A multi-wireless body area network system model based on bidirectional transmission of information and energy is constructed, and two transmission protocols based on time division multiple access are constructed for different network topologies; the multi-wireless body area network system model comprises a pre-arranged radio frequency energy source for wireless energy supply, a plurality of freely movable wireless body area networks, and an AP equipped with an edge server; each wireless body area network comprises a sink node and m sensors located at different positions;

[0010] The receiving and transmitting processes of wireless energy transmission and wireless data transmission are mathematically described, and the energy collection and data transmission performance of each sensor node and the information collection performance of the sink node are analyzed;

[0011] An overall optimization model is constructed for three sub-problems of a transmission protocol selection vector a, a sensor transmission power allocation ratio parameter p ij , and a time slot order P;

[0012] The three sub-problems are alternately optimized and updated to solve the optimal configuration of the multi-wireless body area network system model, and the optimal transmission strategy a * , p * , and P * are obtained.

[0013] Further, the wireless body area networks in the multi-wireless body area network system model are divided into c groups according to the network topology, and there is no inter-network interference between the wireless body area networks in different groups, and optimization is performed according to the groups.

[0014] Further, in the transmission protocol, the entire frame is divided into an uplink phase, a downlink phase, and a transmission phase.

[0015] In the uplink phase, each sensor node transmits position information.

[0016] In the downlink phase, the AP broadcasts the transmission protocol of each wireless body area network in this phase, the transmission time slot order and transmission power allocation ratio of each sensor node, and synchronization information.

[0017] The transmission stage is divided into an energy collection stage and an information transmission stage. In the energy collection stage, the radio frequency energy source broadcasts a radio frequency signal, and the sensor node receives the energy signal and charges;

[0018] The entire information transmission stage is divided into a normal transmission stage and an emergency transmission stage. In the normal transmission stage, each sensor node transmits to the sink node according to a transmission protocol based on time division multiple access technology, and the sink node forwards the data to the AP;

[0019] In the emergency transmission stage, the sensor nodes with remaining data compete for transmission opportunities according to a probability determined by the weight.

[0020] Further, before the transmission stage, the AP predicts the topology of the network and makes a strategy based on the topology; and broadcasts a radio frequency signal carrying strategy information to the sensor nodes using the SWIPT technology, and the sensor nodes collect energy in this stage; in the information transmission stage, the sensor nodes send the collected physiological data to the sink node according to the received transmission strategy information, and the sink node forwards the data to the AP.

[0021] Further, the transmission protocol in the emergency transmission stage is:

[0022] The importance of the node is quantified, and the importance of the node is defined as:

[0023] v=c1*H+F

[0024]

[0025] Where F is a fairness factor; H is the degree of the index deviating from the normal range; c1 is a constant; θ l , θ u represents the maximum and minimum values of the normal range of the index; θ is the number of sample data values;

[0026] The distribution of the importance of the node is represented by a discrete distribution as:

[0027]

[0028] The importance of the node changes with time, and the optimization problem is represented as:

[0029]

[0030] s.t.C1:

[0031] C2:ω i >ω min

[0032] Wherein, denotes the data of the ith sample; P ∈ E denotes the set of consistent distribution; E P denotes expectation; r(ω, θ) denotes loss function; C1 denotes the normalization of weight; C2 denotes the minimum weight constraint of guarantee node;

[0033] The above random optimization problem is converted into a convex optimization problem according to a distributed robust optimization algorithm:

[0034]

[0035] s.t.C1:

[0036] C2: λ ≥ 0

[0037] C3:

[0038] C4: ω i > ω min

[0039] Wherein, λ is the Lagrange multiplier; s i is the introduced variable; is the sample data; ||.|| is the norm; is any ith sample; ω min is the minimum weight constraint; C1 is the constraint of introducing the mirror image; C2 is the constraint of the Lagrange multiplier; C3 is the weight normalization constraint; C4 is the minimum constraint on the weight;

[0040] The optimal weight ω * is obtained by solving the above convex optimization problem.

[0041] 8、Further, the overall optimization model realizes the maximization of the overall transmission rate of the multi-wireless body area network system on the premise of guaranteeing the normal service quality of the sensor node, and the expression of the objective function is as follows:

[0042]

[0043] s.t.C1:

[0044] C2:

[0045] C3:

[0046] C4:

[0047] C5:

[0048] C6:

[0049] where, r1(p ij ) is the throughput under the first transmission protocol; r2(p ij ) is the throughput under the second transmission protocol; g is the channel gain from the AP to the nodes; j is the jth user; M(g) is the users in the same group; a j is the protocol selection vector; T is the vector transpose; ω ij is the node weight; t i is the time allocated to the ith time slot; η is the energy conversion efficiency of the nodes; is the channel gain from the sensor nodes to the sink node, d ij is the ith node of the jth user, s j is the sink node of the jth user; is the channel gain from the AP to the sensor nodes for energy signal, e is the AP transmit energy signal; P j is the set of nodes that transmit in the same time slot as the node; is the channel gain from the sink node to the sensor nodes; is the channel gain from the AP to the sensor nodes for energy signal; d ik is the ith node of the kth user; ρ ik is the power allocation ratio; P s is the power of the AP transmit energy signal; σ 2 is the Gaussian white noise power; B is the bandwidth; r min represents the minimum transmission rate that meets the system quality of service requirements;

[0050] C1, C2 are the constraints that the data transmission must meet the quality of service; C3 is the non-negative constraint and upper bound constraint on the power allocation ratio; C4 constrains the energy consumed by the sensor node to transmit information to be less than the received energy; C5, C6 are the constraints on the protocol selection.

[0051] Further, the optimal configuration of the multi-wireless body area network system model is solved based on the overall optimization model to obtain an optimal transmission strategy, comprising:

[0052] According to the network topology graph, the body area network is grouped according to whether there is body area network interference, and the initial values of the transmission protocol selection vector a, the power allocation ratio parameter ρ and the time slot order P are set;

[0053] The optimal protocol selection a * is calculated by a convex optimization algorithm, and the common parameters are updated; under the new common parameters, the optimal time slot order P * is obtained by using the Hungarian algorithm, and the common parameters are updated; under the new common parameters, the objective function is rewritten as the form of convex function minus convex function, and the optimal sensor transmit power allocation ratio ρ is calculated by using the convex difference algorithm* and update the common parameters;

[0054] Repeat the above algorithm until the difference between two adjacent optimizations is less than a convergence threshold, and obtain the final optimal strategy a * , P * , ρ * .

[0055] Further, the optimal protocol selection a * is calculated by a convex optimization algorithm

[0056] Under the condition that the time slot order P and the power allocation ratio parameter ρ are fixed, the transmission protocol selection sub-problem is expressed as:

[0057]

[0058] s.t.C1:

[0059] C2: a j (i) ∈ {0, 1}

[0060] By calculation:

[0061]

[0062] wherein, P j represents a set of body area networks that have inter-network interference with the body area network j; then:

[0063]

[0064] When , the transmission protocol selection vector a j = (0, 1); when , the transmission protocol selection vector a j = (1, 0).

[0065] Further, the optimal time slot order P * is obtained by using a Hungarian algorithm

[0066] A utility matrix U is established, U ij = αd ij + β ij

[0067] wherein α and β represent the attitudes of the decision maker to the communication quality of important data and the reduction of inter-network interference; d ij represents the distance between two sensor nodes, and w ij represents the weight ratio of the two sensor nodes;

[0068] Subtract all elements in the utility matrix from the maximum element to obtain a new matrix C, c ij = U imax - U ij ;

[0069] Calculate the optimal time slot order P by using the Hungarian algorithm * .

[0070] Further, calculate the optimal sensor transmission power distribution ratio p by using the convex difference algorithm * , comprising:

[0071] Fix a and P as constants, find the maximum value of the total transmission rate R( p) of the multi-wireless body area network system model to optimize the communication performance; split the objective function into each time slot in a single group and optimize respectively:

[0072]

[0073] Rewrite R t into the form of the difference between convex functions and convex functions:

[0074] R t ( p tj ) = Y t ( p tj ) - F t ( p tj ) + D

[0075] Convert the optimization problem into an unconstrained problem in the form of a penalty function;

[0076] Obtain the optimal sensor node transmission power distribution ratio p by using the convex difference algorithm * .

[0077] According to the above technical solution, compared with the prior art, the present application has the following beneficial effects:

[0078] The present application aims at the interference problem between networks under the coexistence of multi-wireless body area networks, and proposes to reduce inter-network interference and improve the network performance of the overall system by controlling different transmission protocols, node transmission order and node transmission power distribution ratio, which can ensure high reliability of information transmission in the multi-wireless body area network information energy bidirectional transmission scene. In addition, the existing subjective judgment weight quantification method depending on the decision maker is improved, and a dynamic weight calculation algorithm based on data is proposed, which more accurately ensures the transmission quality of important nodes. The technology described in the present application can not only be applied in the field of medical health, but also has great practical value in entertainment, health and military fields. BRIEF DESCRIPTION OF DRAWINGS

[0079] In order to make the technical solutions in the embodiments of the present application or the prior art clearer, the accompanying drawings needed in the embodiments or prior art description will be briefly introduced. Obviously, the accompanying drawings in the following description only aim to explain part of the embodiments of the present application, and all other embodiments obtained by those of ordinary skill in the art without creative effort based on the embodiments in the present application shall fall within the protection scope of the present application.

[0080] Figure 1 The flow chart of the resource allocation method in the coexistence of multiple wireless body area networks provided by the present application.

[0081] Figure 2 The structural schematic diagram of the multiple wireless body area network system model provided by the present application.

[0082] Figure 3 The transmission protocol schematic diagram provided by the present application. DETAILED DESCRIPTION

[0083] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort shall fall within the protection scope of the present application.

[0084] As shown in Figure 1 , the present application discloses a resource allocation method in the coexistence of multiple wireless body area networks, comprising the following steps:

[0085] S1, a multiple wireless body area network system model based on bidirectional transmission of information energy is constructed, and two transmission protocols based on time division multiple access are constructed for different network topological structures; the multiple wireless body area network system model comprises a pre-arranged radio frequency energy source for wireless energy supply, a plurality of (N) wireless body area networks which can be freely moved, and an AP (wireless access point) provided with an edge server; each wireless body area network comprises a sink node and m sensors located at different positions; the wireless body area networks in the multiple wireless body area network system model are divided into c groups according to the network topological structure, and there is no inter-network interference between the wireless body area networks in different groups, and optimization is performed according to the groups.

[0086] S2, the transmission and reception process of wireless energy transmission and wireless data transmission is mathematically described, and the energy collection and data transmission performance of each sensor node and the information collection performance of the sink node are analyzed;

[0087] S3, an overall optimization model is constructed for three sub-problems of a transmission protocol selection vector a, a sensor transmission power allocation ratio parameter ρ ij and a time slot order P in the transmission process.

[0088] S4, alternately optimizing and updating the three sub-problems to solve the optimal configuration of the multi-wireless body area network system model, and obtaining an optimal transmission strategy a * , p * , and P * .

[0089] In one embodiment, the transmission protocol in S1 is divided into uplink, downlink and transmission stages for the entire frame.

[0090] In the uplink stage, each sensor node transmits position information; the sensor transmits the position information to the sink node based on the time division multiple access technology, and the sink node forwards the data to the AP, which makes decisions and optimizations after receiving the data.

[0091] In the downlink stage, the AP broadcasts the transmission protocol of each wireless body area network in this stage, the transmission time slot order and transmission power allocation ratio of each sensor node, and synchronization information.

[0092] The transmission stage is divided into an energy collection stage and an information transmission stage. In the energy collection stage, the radio frequency energy source broadcasts a radio frequency signal, and the sensor node receives the energy signal and charges.

[0093] The entire information transmission stage is divided into a normal transmission stage and an emergency transmission stage. In the normal transmission stage, each sensor node transmits the collected physiological data to the sink node according to the transmission protocol based on the time division multiple access technology, and the sink node forwards the data to the AP.

[0094] In the emergency transmission stage, the sensor node with remaining data competes for transmission opportunities according to the probability determined by the weight.

[0095] Before the transmission stage, the AP predicts the topology of the network and makes strategies based on the topology, groups the body area networks in the system according to whether they are within the communication range, and optimizes each group separately; and uses the SWIPT technology to broadcast the radio frequency signal carrying the strategy information to the sensor node, and the sensor node collects energy in this stage; in the information transmission stage, the sensor node sends the collected physiological data to the sink node according to the received transmission strategy information, and the sink node forwards the data to the AP, and designs two transmission protocols based on time division multiple access for different network topologies, the first transmission protocol is that only one node transmits in a time slot in a group, and the second transmission protocol is that only one node transmits in a time slot in a body area network. Both protocols include normal and emergency transmission stages.

[0096] In one embodiment, the overall optimization model maximizes the overall transmission rate of the multi-WBAN system, reduces the inter-network interference and avoids the intra-network interference, with the premise of ensuring the normal service quality of the sensor nodes, and the objective function is expressed as follows:

[0097]

[0098]

[0099] wherein r1(ρ ij ) is the throughput under the first transmission protocol; r2(ρ ij ) is the throughput under the second transmission protocol; g is the channel gain from the AP to the node; j is the jth user; M(g) is the user in the same group; a j is the protocol selection vector; T is the vector transpose; ω ij is the node weight; t i is the time allocated to the ith time slot; η is the energy conversion efficiency of the node; is the channel gain from the sensor node to the sink node, d ij is the ith node of the jth user, s j is the sink node of the jth user; is the channel gain from the AP to the sensor node, e is the AP transmitting energy signal; P j is the set of nodes transmitting in the same time slot as the node; is the channel gain from the sink node to the sensor node; is the channel gain from the AP to the sensor node; d ik is the ith node of the kth user; ρ ik is the power allocation ratio; P s is the power of the AP transmitting energy signal; σ 2 is the Gaussian white noise power; B is the bandwidth; r min represents the minimum transmission rate meeting the system service quality requirement;

[0100] C1 and C2 are the constraints of the data transmission meeting the service quality; C3 is the non-negative constraint and upper bound constraint of the power allocation ratio; C4 constrains that the energy consumed by the sensor node transmitting information cannot be greater than the received energy; C5 and C6 are the constraints of the protocol selection.

[0101] In one embodiment, S4 comprises:

[0102] S41, grouping the body area networks according to the presence or absence of body area network interference according to the network topology map, setting the initial values of the transmission protocol selection vector a, the power allocation ratio parameter ρ and the time slot order P;

[0103] S42, calculate the optimal protocol selection a by a convex optimization algorithm * , and update the common parameters; under the new common parameters, the optimal time slot order P is obtained by using the Hungarian algorithm * , and update the common parameters; under the new common parameters, the objective function is rewritten as the form of convex function minus convex function, and the optimal sensor transmission power distribution ratio p is calculated by using the convex difference algorithm * , and update the common parameters;

[0104] S43, repeat S42 until the difference between adjacent two optimizations is less than the convergence threshold, and obtain the final optimal strategy a * , P * , p * .

[0105] It should be noted that:

[0106] In actual situations, the optimization order of a, p, and P can be changed, and different optimization orders have an impact on the convergence speed. Through simulation, it is found that the convergence speed is the fastest when the protocol selection a is optimized first, then the time slot distribution P is optimized, and finally the power distribution ratio p is optimized, and optimizing the power distribution ratio has the greatest effect on improving the overall communication performance.

[0107] The optimal transmission strategy is to mathematically express the receiving and transmitting process of wireless energy transmission and wireless data transmission in S2, and analyze the energy collection and data transmission performance of each sensor node and the information collection performance of the sink node, which includes:

[0108] (1) Channel model

[0109] The channel model used in the body area network of the application is:

[0110]

[0111] Where d i represents the transmission distance between the sensor node and the sink node, u represents the path loss exponent, and d0 represents the relevant distance.

[0112] The Gamma distribution can be used to describe the channel characteristics between body area networks, so the channel model between body area networks used in the application is:

[0113]

[0114] Where A represents the amplitude of the channel gain, x, k represents the parameters describing the shape and size of the distribution, and Γ(k represents the channel gain distribution; the application uses mean and variance to represent:

[0115] E(x)=kx (4)

[0116] D(x) = kx 2 (5)

[0117] The formula of mean and variance changing with distance is:

[0118] E = ar dB +b,r dB = 20log 10 r (6)

[0119] Wherein, a represents a constant, b represents a constant;

[0120] (2) Transmission process analysis

[0121] In the energy collection stage, the radio frequency energy source broadcasts a radio frequency signal, the sensor node receives the energy signal and charges, so that the node D ij The energy collected is:

[0122]

[0123] Wherein, the transmission power of the radio frequency energy source is P s , η represents the energy collection efficiency of the sensor node, and satisfies 0 < η < 1, Indicates the transmission channel gain from the energy source to the sensor node D ij , t0 is the time length of the transmission time slot allocated by the system to the energy transmission stage.

[0124] In the information transmission stage, the sensor transmits the collected life health information to the sink node in turn through the transmission protocol based on time division multiple access technology, and the transmission power of the sensor node is represented as:

[0125]

[0126] Wherein, P ij d Indicates the transmission power of the sensor node D ij , η represents the energy collection efficiency of the sensor node, and satisfies 0 < η < 1, Indicates the transmission channel gain from the energy source to the sensor node D ij .

[0127] (3) Dynamic weight based on data

[0128] The physical conditions of different monitored persons are different, so the importance of the nodes is also different; even for the same monitored person, the importance of the nodes is different at different time stages. In the medical field, important nodes can more affect the life health of the monitored person, so the application proposes a dynamic weight calculation method based on data rather than subjective judgment, which specifically includes:

[0129] The importance of nodes is quantified, and the importance of a node is defined as follows:

[0130] v = c1*H + F (9)

[0131]

[0132] Where F is the fairness factor; H is the degree to which the indicator deviates from the normal range; c1 is a constant; θ l θ u This represents the maximum and minimum values ​​within the normal range of the indicator; θ represents the sample data values.

[0133] The distribution of node importance can be represented by a discrete distribution as follows:

[0134]

[0135] The importance of nodes changes over time, and the optimization problem can be expressed as minimizing the loss function under the worst-case distribution:

[0136]

[0137] in, Let P represent the data of the i-th sample; P∈E represents the set of distributions that conform to it; E P θ represents the expectation; r(ω,θ) represents the loss function; C1 represents the weight normalization; C2 represents the minimum weight constraint to guarantee the node.

[0138] Based on the distributed robust optimization algorithm, the above stochastic optimization problem is transformed into a convex optimization problem (i.e., the Lagrangian function of the previous problem):

[0139]

[0140]

[0141] Where λ is a Lagrange multiplier; ,S i Variables introduced; For sample data; ||.|| represents the norm; Let ω be any i-th sample; min C1 is the minimum weight constraint; C2 is the constraint for introducing the upper view image; C3 is the Lagrange multiplier constraint; C4 is the weight normalization constraint; C5 is the minimum constraint on the weights.

[0142] The optimal weight ω is obtained by solving the above convex optimization problem using a tool for solving convex optimization problems. * .

[0143] (4) Transmission rate analysis during transmission process

[0144] In the stage of transmitting the life health information from the sensor node to the sink node, node D ij Transmitting data to sink node S j The signal-to-noise ratio of the process is expressed as:

[0145]

[0146] where σ 2 represents the power of the Gaussian white noise with zero mean and variance σ 2 , represents the channel gain between the sensor node D ij and the sink node S j .

[0147] The transmission power of node D ij to the sink node S j is:

[0148]

[0149] The total transmission rate of the system with weights in one transmission frame can be expressed as:

[0150]

[0151]

[0152]

[0153] where r1(ρ ij ) represents the signal-to-noise ratio of the transmission under transmission protocol one, and r2(ρ ij ) represents the signal-to-noise ratio of the transmission under transmission protocol two.

[0154] (5) Model solution

[0155] The smaller the distance between the body area networks in the communication range, the smaller the inter-network interference, and thus the greater the transmission rate. According to the analysis, it can be obtained that the sum of the transmission rates of the body area networks transmitting under transmission protocol one in the same body area network group is proportional to the sum of the distances between the body area networks in the group. The body area networks transmitting under transmission protocol two in the same body area network group do not have inter-network interference because only one node transmits in each time slot. The sum of the transmission rates of the body area networks transmitting under transmission protocol two in the same body area network group does not change with the change of the distances between the body area networks.

[0156] The transmission performance of the system as a whole is improved, and the inter-network interference is reduced by formulating an optimal transmission strategy to maximize the total transmission rate of the system. Since the optimization problem is a polynomial difficult problem, the optimization problem is divided into three optimization sub-problems, which are solved respectively. The optimization sub-problems include: a transmission protocol optimization sub-problem, a time slot design optimization sub-problem and a node transmission power distribution ratio optimization sub-problem.

[0157] Specifically, in S42, the optimal protocol selection a * is calculated by using a convex optimization algorithm.

[0158] When the time slot sequence P and the power distribution ratio parameter p are fixed, the transmission protocol selection sub-problem (solving the optimal transmission protocol) is expressed as:

[0159]

[0160] By calculation, we have:

[0161]

[0162] wherein, P j represents a set of body area networks having inter-network interference with the body area network j; and:

[0163]

[0164] When , the communication performance of the transmission in the second transmission protocol is better than that in the first transmission protocol, and the transmission protocol selection vector a j =(0, 1); when , the performance of the first transmission protocol is better, and the transmission protocol selection vector a j =(1, 0).

[0165] In S42, the optimal time slot sequence P * is obtained by using the Hungarian algorithm, which includes:

[0166] Since the distance between nodes and the weight of the nodes affect the allocation of the time slot sequence, the utility function is introduced in the application: the utility matrix U is established:

[0167] U ij =αd ij +βw ij (22)

[0168] wherein, α and β represent the attitude of the decision maker to guarantee the communication quality of important data and reduce the inter-network interference; d ij represents the distance between two sensor nodes, and w ij represents the weight ratio of two sensor nodes.

[0169] The maximum value element in the utility matrix is subtracted from all elements to obtain a new matrix C, c ij = U imax - U ij ;

[0170] The optimal time slot order P is calculated by using the Hungarian algorithm on the matrix C * .

[0171] In S42, the optimal sensor transmission power distribution ratio p is calculated by using the convex difference algorithm * , comprising:

[0172] Fix a and P as constants, and find the maximum value of the total transmission rate R(p) of the multi-wireless body area network system model to optimize the communication performance; since there is no interference between nodes in different groups or different time slots in the same group, the objective function is divided into each time slot in a single group for optimization:

[0173]

[0174] R t is rewritten in the form of the difference between convex functions and convex functions:

[0175] R t (p tj ) = Y t (p tj ) - F t (p tj ) + D (24)

[0176]

[0177]

[0178]

[0179]

[0180] Since the convex difference algorithm is applied to an unconstrained problem, the functions of each constraint condition and p tj have a linear relationship, so the optimization problem is converted into an unconstrained problem in the form of a penalty function;

[0181]

[0182] Wherein, u(i) is a non-negative penalty factor, h i + (p tj ) = max[0, h i (p tj )].

[0183]

[0184]

[0185] The optimal sensor node transmit power allocation ratio p is obtained by a convex difference algorithm * .

[0186] It should be further explained that:

[0187] The penalty function is accurate when the penalty factor is greater than a certain lower bound, and the lower bound is related to the Lagrange multiplier.

[0188] The solving process using the convex difference iterative algorithm is as follows:

[0189] First, an initial point p0 is set, and the following convex problem is solved:

[0190]

[0191] where <,> represents the inner product, denotes the partial derivative, and the optimal solution of the above formula is obtained

[0192] Second, the descending direction is: The step size is l k , wherein the step size satisfies:

[0193]

[0194] The joint optimization is achieved by an alternating update algorithm. The optimal value of another optimization variable is solved and the common parameter is updated while the other two optimization variables are fixed. The other variables are optimized in the same way. An alternating optimization is achieved every time each variable completes an optimization. The above is repeated until a preset condition is reached or a preset number of iterations is reached, and the final joint optimization result is obtained.

[0195] According to the above discussion, the dynamic weight based on data can be solved by using a distributed robust optimization algorithm, the transmission protocol sub-problem can be solved by combining the characteristics of the protocol and convex optimization, the time slot order optimization can be solved by establishing an utility matrix and using the Hungarian algorithm, and the optimal power allocation ratio can be obtained by using a convex difference algorithm combined with a penalty function. Based on the above, each variable can be alternately optimized and updated. The main idea is as follows: the optimal value of another optimization variable is solved and the common parameter is updated while the other two optimization variables are fixed. The other variables are optimized in the same way. An alternating optimization is achieved every time each variable completes an optimization. The above is repeated until a preset condition is reached or a preset number of iterations is reached, and the final joint optimization result is obtained.

[0196] The various embodiments described in this specification are presented by way of example, and each embodiment is presented with the understanding that it will not limit the present application to that embodiment alone. Each embodiment is presented in its own right, and the various embodiments are not mutually exclusive, but can be combined with each other. The various embodiments can be implemented in any combination or sub-combination thereof.

[0197] The above description of disclosed embodiments is intended to be illustrative and not restrictive. Many modifications of the embodiments as described will become apparent to those skilled in the art without departing from the spirit of the application as defined by the appended claims. The scope of the application should therefore be determined not with reference to the above description, but instead should be given its broadest interpretation consistent with the principles and novel features disclosed herein.

Claims

1. A method for resource allocation in a coexistence scenario of multiple wireless body area networks, characterized in that, The method comprises the following steps: A multi-wireless body area network system model based on information energy bidirectional transmission is constructed, and two transmission protocols based on time division multiple access are constructed for different network topologies; the multi-wireless body area network system model comprises a pre-arranged radio frequency energy source for wireless energy supply, a plurality of freely movable wireless body area networks and an AP provided with an edge server; each wireless body area network comprises a sink node and m sensors located at different positions; The receiving and transmitting processes of wireless energy transmission and wireless data transmission are mathematically expressed, and the energy collection and data transmission performance of each sensor node and the information collection performance of the sink node are analyzed; Aiming at the transmission protocol selection vector a, sensor transmission power allocation ratio parameter p in the transmission process ij And time slot sequence P three sub-problems to build the overall optimization model; The three sub-problems are alternately optimized and updated to solve the optimal configuration of the multi-wireless body area network system model, and an optimal transmission strategy a is obtained * , p * , and P * ; Based on the overall optimization model, the optimal configuration of the multi-wireless body area network system model is solved, and an optimal transmission strategy is obtained, comprising: According to a network topology graph, the body area networks are grouped according to whether there is inter-body area network interference, and initial values of a transmission protocol selection vector a, a power allocation ratio parameter p and a time slot order P are set; The optimal protocol selection a is calculated by a convex optimization algorithm * , and the common parameters are updated; under the new common parameters, the optimal time slot order P is obtained by using the Hungarian algorithm * , and the common parameters are updated; under the new common parameters, the target function is rewritten into the form of convex function minus convex function, and the optimal sensor transmission power distribution ratio p is calculated by using the convex difference algorithm * , and the common parameters are updated; The above algorithm is repeated until the difference between two adjacent optimizations is less than a convergence threshold, obtaining the final optimal strategy a * , P * , p * ; The optimal protocol selection a is calculated by a convex optimization algorithm * comprising: Under the condition that the time slot order P and the power allocation ratio parameter p are fixed, the transmission protocol selection sub-problem is represented as follows: C2: a j (i) ∈ {0, 1} Wherein, T is a vector transposition; Through calculation, the following is obtained: where denotes the channel gain from the sensor node d ij to the sink node s j ; d ij denotes the i-th sensor node of the j-th user; s j denotes the sink node of the j-th user; denotes the channel gain from the access node to the sensor node d ik ; d ik denotes the i-th sensor node of the k-th user; p ik denotes the power allocation ratio of the i-th sensor node of the k-th user; η is the energy conversion efficiency of the node; σ 2 is the Gaussian white noise power; P j denotes the set of body area networks that have an inter-network interference with body area network j; then: When the transmission protocol selection vector a j = (0,1); when the transmission protocol selection vector a j = (1,0); The optimal time slot sequence P is obtained by using the Hungarian algorithm * comprising: A utility matrix U, U ij = αd ij + βw ij wherein a, β represent the attitude of decision maker for ensuring the communication quality of important data and reducing inter-network interference; d ij denotes the distance between two sensor nodes, w ij denotes the weight ratio of two sensor nodes; Subtracting the maximum element in the utility matrix from all elements, we get a new matrix C, c ij = U imax - U ij ; The optimal time slot order P is calculated using the Hungarian algorithm * ; The optimal sensor transmit power distribution ratio p is calculated by using the convex difference algorithm * comprising: The maximum value of R(p) of the total transmission rate of the multi-wireless body area network system model is solved by fixing a and P as constants, so as to optimize the communication performance; the objective function is divided into each time slot in a single group for optimization: R t Rewrite as the difference of convex functions: R t (ρ tj )=Y t (ρ tj )-F t (ρ tj )+D The optimization problem is converted into an unconstrained problem in the form of a penalty function; The optimal sensor node transmit power allocation ratio p is obtained by the convex difference algorithm * .

2. The method of claim 1, wherein, According to the network topology structure, the wireless body area networks in the multi-wireless body area network system model are divided into c groups, and there is no inter-network interference between the wireless body area networks in different groups, and the wireless body area networks are optimized according to the groups. 3.The method of claim 1, wherein, In the transmission protocol, for the whole frame, it is divided into uplink, downlink and transmission stages; In the uplink stage, each sensor node transmits position information; In the downlink stage, the AP broadcasts the transmission protocol of each wireless body area network in this stage, the transmission time slot order and transmission power allocation ratio of each sensor node and synchronization information; The transmission stage is divided into an energy collection stage and an information transmission stage, in the energy collection stage, the radio frequency energy source broadcasts a radio frequency signal, and the sensor nodes receive the energy signal and charge; The whole information transmission stage is divided into a normal transmission stage and an emergency transmission stage, in the normal transmission stage, each sensor node transmits to the sink node according to the transmission protocol based on the time division multiple access technology, and the sink node forwards to the AP; In the emergency transmission stage, the sensor nodes with remaining data compete for transmission opportunities according to the probability determined by the weight.

4. The method of claim 3, wherein, Before the transmission stage, the AP predicts the topology structure of the network and formulates a strategy based on the topology structure; and the radio frequency signal carrying the strategy information is broadcast to the sensor nodes by using the SWIPT technology, and the sensor nodes collect energy in this stage; In the information transmission stage, the sensor nodes transmit the collected physiological data to the sink node according to the received transmission strategy information, and the sink node forwards the data to the AP.

5. The method of claim 3, wherein, The transmission protocol in the emergency transmission stage is as follows: The importance of the node is quantified, and the importance of the node is defined as: v=c1*H+F where F is a fairness factor; H is a degree of departure from a normal range of the index; c1 is a constant; θ l , θ u denotes maximum and minimum values of a normal range of the index; θ is a sample data value; The distribution of the importance of the node is represented by a discrete distribution as follows: The importance of the node changes with time, and the optimization problem is represented as follows: C2: ω i >ω min wherein, represents data of the i-th sample; P E represents a set of distributions that fit; E P represents expectation; r(ω, θ) represents a loss function; C1 represents a normalization of weights; C2 represents a constraint to guarantee the minimum weight of a node. According to the distributed robust optimization algorithm, the random optimization problem is converted into a convex optimization problem: C2: λ ≥ 0 C4: ω i >ω min where λ is a Lagrange multiplier; s i are introduced variables; is sample data; ||.|| is a norm; is an arbitrary ith sample; ω min is a minimum weight constraint; C1 is a constraint to introduce a mirror image; C2 is a Lagrange multiplier constraint; C3 is a weight normalization constraint; C4 is a minimum constraint on the weight; The optimal weights ω are obtained by solving the above convex optimization problem * .

6. The method of claim 1, wherein, The overall optimization model takes the premise of ensuring the normal service quality of the sensor node to realize the maximization of the overall transmission rate of the multi-WBAN system, and the expression of the objective function is as follows: where r1(p ij ) is the throughput under the first transmission protocol; r2(p ij ) is the throughput under the second transmission protocol; g is the channel gain from the AP to the node; j is the jth user; M(g) is the users in the same group; a j is the protocol selection vector; ω ij is the node weight; t i is the time allocated to the ith time slot; is the channel gain from the sink node to the sensor node; P s is the power of the AP transmit energy signal; B is the bandwidth; r min denotes the minimum transmission rate that satisfies the system quality of service requirement; C1 and C2 are constraints of data transmission to meet the service quality; C3 is a non-negative constraint and an upper bound constraint on the power allocation ratio; C4 restricts that the energy consumed by the sensor node to transmit information cannot be greater than the received energy; C5 and C6 are constraints for protocol selection.

Images