CN109362128B - Method for improving frequency spectrum efficiency of edge wireless terminal of cellular weak coverage area - Google Patents

Abstract

The invention discloses a method for improving the frequency spectrum efficiency of a wireless terminal at the edge of a cellular weak coverage area, and particularly relates to how to combine free frequency spectrum supply and radio frequency energy compensation to construct an excitation framework so as to promote cooperation among wireless terminals and achieve the aim of improving the frequency spectrum efficiency of the wireless terminal at the edge of the cellular weak coverage area. By recruiting devices with energy storage advantages and geographic location advantages as charging nodes instead of relying on the SBS to provide radio frequency energy, the invention can obtain a large amount of energy by the SVAP rather than waste on a transmission path.

Description

Method for improving frequency spectrum efficiency of edge wireless terminal of cellular weak coverage area

Technical Field

The invention belongs to the technical field of wireless network performance optimization, relates to a method for improving the frequency spectrum efficiency of a wireless terminal at the edge of a cellular weak coverage area, and particularly relates to how to combine free frequency spectrum supply and radio frequency energy compensation to construct an excitation framework so as to promote cooperation among wireless terminals and achieve the aim of improving the frequency spectrum efficiency of the wireless terminal at the edge of the cellular weak coverage area.

Background

According to the report of the literature [1], with the explosive development of the internet, the peak value of the internet traffic is increased faster than the average rate, and the difference between the two is in an increasing trend. The traffic for tablet and laptop, smart phone, machine to machine communication, consumer internet of things applications will exceed fixed IP traffic. In addition to enhanced mobile broadband applications, this growing primary driving force will come from the growing demand of the consumer wearable device market. In 2017, the market value of wearable medical devices worldwide reaches about $ 52 billion, and is expected to grow to $ 135 billion by 2022.

An effective approach for mobile operators to cope with explosive traffic demands is to deploy heterogeneous cellular networks in ultra-dense fashion. For example, a femtocell may be deployed in each macrocell with an overlay, and a picocell and a femtocell may be deployed in each macrocell with an overlay. In fact, mobile operators, despite building such ultra-dense heterogeneous cellular networks in areas of emphasis they consider, have difficulty in achieving ubiquitous and seamless ultra-dense coverage. For example, for areas such as gymnasiums, world expo centers, and suburban industrial parks, mobile operators consider that traffic demand is not high most of the time and will not typically build cellular networks at peak traffic demand. However, activities such as civil sporting events, festival celebrations, large commercial exchanges, industrial equipment exposition, etc. may occasionally be held in these locations, resulting in short term ultra high traffic demands, where a cellular weak coverage area is necessarily formed. In such an area, if a wireless terminal located at the edge of the cellular coverage area applies for a spectrum resource, it is difficult to make full use of the spectrum resource, which is the reason: on one hand, in such a cellular weak coverage area, the edge wireless terminal is usually far away from the base station, and a communication link between the edge wireless terminal and the base station may also be blocked by an obstacle, so that the quality of a communication channel is usually poor, and even if the edge wireless terminal uses the maximum transmission power, it may be difficult to overcome signal attenuation caused by the poor channel, so that it is difficult to achieve a data rate required by an application experience; on the other hand, higher transmission power not only consumes own battery energy more quickly, but also causes greater interference to users with the same frequency channel in the adjacent cellular coverage area, and influences the application experience of the users.

Disclosure of Invention

In view of the defects of the prior art, the invention aims to provide a method for improving the spectrum efficiency of an edge wireless terminal in a cellular weak coverage area, which is used for improving the spectrum efficiency of the edge wireless terminal in the cellular weak coverage area by constructing an excitation architecture based on free spectrum supply and radio frequency energy compensation.

The method for improving the frequency spectrum efficiency of the wireless terminal at the edge of the cellular weak coverage area comprises the following steps:

(1) planning a plurality of reference points in a cellular weak coverage area, and selecting a wireless terminal with energy storage advantages and geographic position advantages as a Virtual Access Point (SVAP) for each reference Point according to a preset standard;

(2) each SVAP is responsible for recruiting wireless terminals with Energy storage advantages and geographic position advantages to serve as Small-cell Energy provider roles (SEPs), each SVAP independently selects a group of SEPs to provide charging service for the SVAP, and the SEPs also obtain the frequency spectrum of a free access base station as the return for providing the charging service;

(3) each SVAP is associated with a group of edge wireless terminals so as to provide relay forwarding service for the edge wireless terminals conveniently, and a part of the frequency spectrum resources of the edge wireless terminals is exchanged according to the contribution of forwarding data and used as a free access base station;

(4) a Small cell Base Station (SBS) gives a division ratio of initial spectrum resources, a game process is started, unit spectrum resources obtained by each edge wireless terminal are divided into three parts, one part is used for sending data of the edge wireless terminal, and the other two parts are respectively used for rewarding SVAP (singular value decomposition) for helping the edge wireless terminal to forward the data and a group of SEPs (secure applications for charging the SVAP);

(5) each SEP in each group of SEPs determines the power for providing charging through a non-cooperative game process; after the charging power of all SEPs is obtained, each SVAP determines the power of the SVAP for forwarding data for the edge wireless terminal through a non-cooperative game process, and the SBS updates the division ratio of the frequency spectrum resources based on the feedback results of the SVAPs and the SEPs;

(6) and (5) repeating the step (5) until the variation value of the partition ratio of the unit spectrum resources is smaller than a preset threshold value, obtaining a Nash equilibrium point of the Stackelberg game, and obtaining a spectrum resource matching scheme with the SEP and SVAP maximized utility.

The steps (1) to (3) of the invention relate to the role construction of the excitation architecture, including the SBS electing SVAPs, and each elected SVAP electing a group of SEPs and associating a group of edge wireless terminals.

In one embodiment, in step (1), the algorithm process of the SBS voting SVAPs is as follows:

1. election operations performed by SBS s:

step 1.0: initializing set S_sIf the SVAPs are empty, the selected SVAPs are stored, and then the step 1.1 is carried out;

step 1.1: determining the number of SVAPs needing to be elected, determining the positions of reference points according to the number and the distribution condition of the areas needing to be enhanced, wherein the number of the reference points is the same as the number of the SVAPs (for example, 8), and then entering step 1.2;

step 1.2: if the reference point has not been traversed, then an SVAP election packet is broadcast at maximum transmit power (e.g., 1000 milliwatts) for an unprocessed reference point, and a time interval (e.g., Δ) is set₁100 ms) to wait for a response packet to return, step 1.3 is entered; otherwise, the operation executed by the SBS s is ended;

step 1.3: if the time interval Δ₁If the consumption is not finished, giving up the wireless terminal meeting the 5 SVAP election conditions in the previous round if the received response packet meets the 5 SVAP election conditions, reserving the wireless terminal meeting the 5 SVAP election conditions at present, and continuously waiting for the next response packet; otherwise, sending SVAP confirmation packet to the reserved wireless terminal, determining the SVAP confirmation packet as the SVAP of the currently investigated reference point, and storing the SVAP confirmation packet in the set S_sThen returning to the step 1.2;

2. the wireless terminal participating in SVAP election executes the following operations:

step 2.0: if an SVAP election packet aiming at a certain reference point is received and the first 4 SVAP election conditions are met, sending a response packet to SBS s, and entering the step 2.1; otherwise, quitting;

step 2.1: if receiving the SVAP confirmation packet from the SBS, marking the identity of the SVAP as the SVAP of a certain reference point, and then finishing the operation; otherwise, waiting for SVAP election packages aiming at other reference points or exiting in time-out.

In one embodiment, in step (2), the algorithm process of the SVAP electing the SEP is as follows:

3. election operations performed by SVAP v:

step 3.0: initializing the set R_vIf empty, to store the selected SEPs, then go to step 3.1;

step 3.1 if set R_vDoes not reach a preset value (e.g., 6), an SEP election packet is broadcast at a maximum transmit power (e.g., 200 milliwatts), and a time interval (e.g., Δ) is set₂100 ms) to wait for a response packet to return, then go to step 3.2; otherwise, the election operation executed by the SVAP v is finished;

step 3.2: if the time interval Δ₂If the consumption is not finished, the received response packet is discarded for the wireless terminal meeting the 5 SEP election conditions in the previous round if the received response packet meets the 5 SEP election conditions, the wireless terminal meeting the 5 SEP election conditions at present is reserved, and the next response packet is continuously waited; otherwise, sending SEP confirmation packet to reserved wireless terminal and storing it in set R_vThen return to step 3.1.

4. The wireless terminal participating in SEP election executes the following operations:

step 4.0: if an SEP election packet of a certain SVAP (e.g., v) is received and the first 4 SEP election conditions are met, sending a response packet to the SVAP v, and entering the step 4.1; otherwise, quitting;

step 4.1: if an SEP confirmation packet from a certain SVAP (for example, v) is received, marking the SEP with the identity of the SEP as the SVAP v, and then finishing the operation; otherwise, waiting for the SEP election package of the SVAP or exiting in time.

In one embodiment, in step (3), the algorithm process of associating the SVAP with a group of edge wireless terminals is as follows:

5. response association operations performed by SVAP v:

step 5.0: initializing set U_vEmpty, ready to store the associated edge wireless terminal, go to step 5.1;

step 5.1: if an association request packet is received from an edge wireless terminal, a feedback is made at the maximum transmit power (e.g., 200 mW)An associated response packet, and then a time interval (e.g., Δ) is set₃100 ms) to wait for an association acknowledgement packet to return, and then to step 5.2;

step 5.2: if the time interval Δ₃If not, the sender is stored in the set U according to the received associated acknowledgement packet_vContinuing to wait for the next association confirmation packet; otherwise, the response association operation performed by the SVAP v ends.

6. Performing operations of the associated edge wireless terminal:

step 6.0: broadcasting an association request packet at a maximum transmit power (e.g., 200 milliwatts) and then setting a time interval (e.g., Δ)₄100 ms) to wait for an association response packet to return, and then go to step 6.1;

step 6.1: if the time interval Δ₄If not, reserving the sender of the association response packet with the strongest current received signal aiming at each feedback association response packet, and continuously waiting for the next association response packet; otherwise, sending an association confirmation packet to the reserved SVAP, and ending the association operation.

The wireless terminal selected as SVAP must meet the following conditions: 1) common wireless interfaces (such as LTE, WiFi, bluetooth) are equipped; 2) ensuring that the mobile phone does not move within a given time; 3) there is a need to communicate with a base station but the required spectrum resources are not currently available; 4) having the capability of capturing radio frequency energy; 5) among all wireless terminals satisfying the first four conditions, the one closest to a certain reference point.

The wireless terminal selected as the SEP must meet the following conditions: 1) ensuring that the mobile phone does not move within a given time; 2) there is a need to communicate with a base station but the required spectrum resources are not currently available; 3) the energy reserve is not less than a given threshold; 4) having a direct communication link with a SVAP; 5) among all wireless terminals satisfying the first four conditions, the one closest to the SVAP.

The steps (4) to (6) of the invention relate to the forming process of an excitation mechanism, and comprise a non-cooperative game process inside each group of SEPs, a non-cooperative game process among SVAPs, and a Stackelberg game process formed by the interaction of SBS, SVAP, SEP and other different roles.

In a specific embodiment, the construction of the excitation mechanism among SBS, SEP and SVAP is completed based on a binary search concept algorithm (min, max), and the specific process is as follows: the SBS gives an initial spectrum resource allocation ratio and starts a game process, a power division ratio is determined inside each set of SEPs through a non-cooperative game process, power division ratios are determined among the SVAPs through a non-cooperative game process, and the SBS determines whether to perform a next round of Stackelberg game process according to feedback results of the SVAPs and the SEPs.

In one embodiment, the specific process of the algorithm BinarySearch (min, max) is as follows:

step 7.0: setting two variables max and min which are respectively used for storing upper and lower boundary values of a reduced half search range, using the initial max and min of the value specified by the algorithm caller, and then entering step 7.1;

step 7.1: judging whether the absolute value of the difference value of max and min is larger than a threshold epsilon, if so, storing the value of (max + min)/2 by using a variable p, and entering the step 7.2; otherwise, the algorithm is ended, wherein epsilon is 0.1% of the difference between the upper and lower boundary values of the initial search range;

step 7.2: judging whether the utility function value with p-epsilon as an independent variable is larger than the utility function value with p as an independent variable; if yes, assigning the value of p to max; otherwise, assigning the value of p to min; then returns to step 7.1 where the utility function is estimated according to equation (18).

In the following algorithm, an SBS (styrene-butadiene-styrene) gives an initial division ratio of a unit resource block W, and then each SEP (secure operating platform) in each group of SEPs determines the power for providing charging through a non-cooperative game process; after the charging power of all SEPs is obtained, each SVAP determines the power of the SVAP for forwarding data for the edge wireless terminal through a non-cooperative game process; finally, the SBS s updates the division ratio of the unit resource block W based on the feedback results of the SVAPs and the SEPs; and repeating the steps until the variation value of the division ratio of the unit frequency spectrum resources is smaller than a preset threshold value, and if the preset threshold value is small enough, obtaining the Nash equilibrium point of the Stackelberg game can be ensured.

In one embodiment, in step (4), the SBS initiates the Stackelberg game process by:

step 8.0: xi is reduced_eAnd xi_vAre all initialized to 0.2 and proceed to step 8.1 where ξ_eA partition ratio value representing a spectrum resource W, the partial resource being used to excite a group of SEPs, the group of SEPs providing a charging service for a certain SVAP, and the SVAP providing a data forwarding service for an owner (i.e., a certain edge wireless terminal) of the spectrum resource W; xi_vRepresenting a division ratio of the spectrum resource W, wherein the part of the resource can be used for rewarding the SVAP providing the data forwarding service for the owner of the spectrum resource W theoretically, but is managed by the SBS s uniformly to encourage the SBS to elect a group of the SVAPs, namely forming a competition relationship, and the contribution larger person can obtain more free use of the part of the resource;

step 8.1: a variable C_eAnd C_vInitializing the SEPs and the SVAPs to be respectively stored to 0, and entering the step 8.2;

step 8.2: for each SEP (e.g., e), power (e.g., to charge its associated SVAP (e.g., v)) is initialized

) Is composed of

Initialize its power to send its data to the SBSs (e.g., power to send its data to the SBSs

) Is composed of

Accumulating the number of SEPs and storing in C_eStep 8.3 is entered, where,

the maximum transmitting power of SEP e is 200 milliwatts;

step 8.3: for each SVAP (e.g., v), initializing its associationThe edge wireless terminal of (2) forwards the power of the data (e.g. power of the data)

) Is composed of

Initializing its power to send its data to SBS s (e.g. power of SBS s)

) Is composed of

Accumulating the number of SVAP and storing in C_vStep 8.4 is entered, where,

the maximum transmitting power of the SVAP v is 200 milliwatts;

step 8.4: calculating a utility function value according to a formula (14), storing the utility function value in a variable mu expression, and entering a step 8.5;

step 8.5: initializing variables

Is C_v/C_eInitialization of

Is 1/C_vThe process proceeds to step 8.6, where,

representing the fraction of SEP e contribution to charging SVAP v,

representing the contribution ratio of the SVAP v to the SBS s for the associated edge wireless terminal to forward data;

step 8.6: SBS s broadcast ξ to all SEPs and SVAPs within the served small cell area_e，ξ_v，

Four parameter messages.

In one embodiment, in the step (5), the specific steps are as follows:

(A) non-cooperative gaming processes within each set of SEPs:

9. operation performed by the wireless terminal e acting as the SEP:

step 9.0: if receiving the message containing xi sent from SBS s_eAnd

in the message (2), the value of min is assigned to 0, and the value of max is assigned to 0

Entering step 9.1;

step 9.1: invoke the algorithm BinarySearch (min, max) to obtain power

Will be provided with

Is assigned a value of

And

the difference between them

And

sending the value of (c) to SBS s, and entering step 9.2;

step 9.2: if the end packet sent by the SBS s is not received, the following operations are repeatedly executed: if received from SBS s transmission

The algorithm is invokedBinarySearch (min, max) to obtain power

Will be provided with

Is updated to

And

the difference between them

And

sends the value of (d) to SBS s and then returns to step 9.2; otherwise, go to step 9.3;

step 9.3: if an end packet is received from the SBS s, the SEP e will be

And

the value of (c) is sent to its associated SVAP v and the algorithm is ended; otherwise, return to step 9.2.

10. Responses performed by SBS s for each SEP (e.g.:

step 10.0: if an inclusion from any SEP (e.g. e) is received

And

if the received value is different from the corresponding value of the SEP e, the newly received value is used for replacing the original value; otherwise, accumulating the number of SEPs with unchanged power division values, and entering the step 10.1;

step 10.1: if at least one SEP in the received messages from SEPs updates the power division value, calculating according to formula (7)

And broadcast an include

To all SEPs. Entering step 10.2;

step 10.2: if the number of the SEPs with unchanged power division values reaches the maximum value of the system (which indicates that no SEP updates the power division values), broadcasting an end packet to all SEPs; otherwise, return to step 10.0.

(B) Non-cooperative gaming process between SVAPs:

11. the wireless terminal v acting as SVAP performs the operations of:

step 11.0: if receiving the signal containing xi sent by SBS s_vAnd

and all related SEPs from the message

And

in the message (2), the value of min is assigned to 0, and the value of max is assigned to 0

Entering step 11.1;

step 11.1: invoke the algorithm BinarySearch (min, max) to obtain power

Will be provided with

Is assigned a value of

And

the difference between them

And

sending the value of (c) to SBS s, and entering step 11.2;

step 11.2: if the end packet sent by the SBS s is not received, the following operations are repeatedly executed: if received from SBS s transmission

The algorithm BinarySearch (min, max) is invoked to obtain power

Will be provided with

Is updated to

And

the difference between them

And

sends the value of (d) to SBS s and then returns to step 11.2; otherwise, go to step 11.3;

step 11.3: if receiving the end packet sent by SBS s, ending the algorithm; otherwise, return to step 11.2.

12. Responses performed by SBS s for each SVAP (e.g., v):

step 12.0: if a packet is received from any of the SVAPs (e.g., v)

And

if the received value is different from the original value, the newly received value is used to replace the original value; otherwise, accumulating the number of SVAPs with unchanged power division values, and entering step 12.1;

step 12.1: if at least one of the SVAPs in the received messages from the SVAPs updates the power division value, then the calculation is made according to equation (12)

And broadcast an include

To all SVAPs. Entering step 12.2;

step 12.2: if the number of the SVAPs with unchanged power division values reaches the system maximum value (which indicates that none of the SVAPs updates the power division values), broadcasting an end packet to all the SVAPs; otherwise, return to step 12.0.

In one embodiment, in step (5), the SBS initiates the Stackelberg game process, with the following specific steps:

(C) the SBS updates the division ratio of the frequency spectrum resources based on the feedback results of the SVAPs and the SEPs:

13. SBS s perform Stackelberg game operations:

step 13.0: if the end packet has been sent to all the SVAPs, assigning the value of min to 0, assigning the value of max to 1, and entering step 13.1;

step 13.1: let xi_eKeeping the current value unchanged, call the algorithm BinarySearch (min, max) to get the new ξ_vAnd entering step 13.2;

step 13.2: let xi_vThe value determined in step 13.1 is kept unchanged and calledThe algorithm BinarySearch (min, max) to obtain the new xi_eAnd entering step 13.3;

step 13.3: with the current newest xi_eAnd xi_vThe value of (d) is an input parameter, the utility function value is calculated by formula (14) and stored in the variable mu' representation, and step 13.4 is entered;

step 13.4: judging whether the mu 'is more than mu or not, if so, assigning the mu as mu', and returning to the step 8.5; otherwise, the algorithm is ended, wherein the initial value of μ is obtained by step 8.4.

The method for improving the spectrum efficiency of the edge wireless terminal in the cellular weak coverage area provided by the invention can avoid the long-distance communication between the edge wireless terminal and the base station by selecting the wireless terminal with the advantages of energy reserve and geographic position as the virtual access point SVAP, and the wireless terminal is replaced by a path consisting of two short-distance links from the edge wireless terminal to the SVAP and from the SVAP to the base station. Therefore, the attenuation degree of the signals transmitted by the edge wireless terminal when reaching the base station can be greatly reduced, so that the edge wireless terminal can meet the data rate requirement required by the user application experience by using lower transmission power and less spectrum resources. The saved spectrum resources may be used to reward the contribution made by the wireless terminal acting as a virtual access point. If a wireless terminal wishes to access the internet through a base station, but temporarily does not apply for the spectrum resources required for communication, there is an incentive to act as a virtual access point. While its energy reserve is sufficient and geographical location is superior, it also makes it more conditional to act as a virtual access point. The virtual access point is responsible for forwarding data of the edge wireless terminal and sending data of the virtual access point, so that even abundant energy reserves can be consumed quickly. There are wireless terminals that have similar motivations as the virtual access point but that fail to become virtual nodes, it may be considered to recruit these wireless terminals to serve the role of small cell energy providers (SEPs) to provide energy supplementary services for the virtual access point in exchange for free access to the communication spectrum resources of the base station. The invention constructs an excitation architecture to realize effective cooperation and mutual profit and win-win of devices with roles of edge wireless terminals, virtual access points (SVAP), wireless terminals (SEP) providing charging service and the like, so that the frequency spectrum resources of the edge wireless terminals are efficiently utilized.

Compared with the prior art, the invention has the beneficial technical effects that:

1) the method for improving the frequency spectrum efficiency of the edge wireless terminal of the cellular weak coverage area provided by the invention has the advantages that on the premise of ensuring the application experience quality of cellular edge equipment users, the rest part of the unit resource block can be coordinated by the SBS as reward resources, so that the problem of low frequency spectrum resource utilization rate of the cellular edge wireless terminal covering the weak area is solved, and the frequency spectrum efficiency of the unit resource block of the edge wireless terminal is improved.

2) In the method for improving the spectrum efficiency of the wireless terminal at the edge of the cellular weak coverage area, wireless interfaces such as WiFi and Bluetooth and the like become conventional configurations of wireless communication equipment, and are combined with the cellular interface and integrated into the solution design to fully utilize the non-overlapping part of the resource blocks.

3) According to the method for improving the spectrum efficiency of the wireless terminal at the edge of the cellular weak coverage area, the device with the energy storage advantage and the geographic position advantage is recruited as the charging node instead of depending on the SBS to provide the radio frequency energy, and a large amount of energy can be acquired by the SVAP instead of being wasted on a transmission path.

4) The method for improving the spectrum efficiency of the edge wireless terminal of the cellular weak coverage area provides an excitation framework based on dynamic charging, and models the problem of resource block division of the edge equipment of the cellular network into a Stackelberg game problem, and the response strategy (namely Stackelberg Nash equilibrium) of all participants can be obtained by running the provided algorithm set.

Drawings

Fig. 1 is a flow chart of a method for improving the spectrum efficiency of a wireless terminal at the edge of a cellular weak coverage area according to the invention.

Fig. 2 is a graph of the average spectral efficiency of edge wireless terminals in a cellular weak coverage area as a function of the number of nodes in a fixed area.

Fig. 3 is a graph of the average spectral efficiency of edge wireless terminals in a cellular weak coverage area as a function of channel noise power in a fixed area.

Figure 4 is a plot of average spectral efficiency of small cell virtual access points and small cell energy providers as a function of the number of nodes in a fixed area.

Figure 5 is a plot of average spectral efficiency of small cell virtual access points and small cell energy providers as a function of channel noise power in a fixed area.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, rather than all embodiments, and all other embodiments obtained by those skilled in the art without any creative work based on the embodiments of the present invention belong to the protection scope of the present invention.

The invention will be further illustrated with reference to the following specific examples and the accompanying drawings:

the method for improving the spectrum efficiency of the edge wireless terminal of the cellular weak coverage area, as shown in fig. 1, comprises the following steps:

(1) planning a plurality of reference Points in a cellular weak coverage area, and selecting a wireless terminal with energy storage advantages and geographic position advantages as a Virtual Access point (SVAPs) for each reference point according to a preset standard;

(2) each SVAP is responsible for recruiting wireless terminals with Energy storage advantages and geographic position advantages to serve as Small-cell Energy provider roles (SEPs), each SVAP independently selects a group of SEPs to provide charging service for the SVAP, and the SEPs also obtain the frequency spectrum of a free access base station as the return for providing the charging service;

(3) each SVAP is associated with a group of edge wireless terminals so as to provide relay forwarding service for the edge wireless terminals conveniently, and a part of the spectrum resources of the edge wireless terminals is exchanged according to the contribution of forwarding data and used as a free access base station, the edge wireless terminals can forward the data to SVAPs by using a Bluetooth technology, and the rest part of the spectrum resources can be used by the SVAPs to forward the data of the edge wireless terminals to the base station;

(4) the SBS gives a division ratio of initial spectrum resources, starts a game process, and divides unit spectrum resources obtained by each edge wireless terminal into three parts, wherein one part is used for sending data of the edge wireless terminal, and the other two parts are respectively used for rewarding SVAP helping the edge wireless terminal to forward the data and a group of SEPs for charging the SVAP;

(5) each SEP in each group of SEPs determines the power for providing charging through a non-cooperative game process; after the charging power of all SEPs is obtained, each SVAP determines the power of the SVAP for forwarding data for the edge wireless terminal through a non-cooperative game process, and the SBS updates the division ratio of the frequency spectrum resources based on the feedback results of the SVAPs and the SEPs;

(6) and (5) repeating the step (5) until the variation value of the partition ratio of the unit frequency spectrum resources is smaller than a preset threshold value, and if the preset threshold value is small enough, obtaining a Nash equilibrium point of the Stackelberg game can be ensured, so that a frequency spectrum resource matching scheme with the maximized SEP and SVAP utilities is obtained.

For each SEP charging a particular SVAP, the free spectrum share it can allocate depends on its contribution to the charging of that SVAP. If the SEP charges the SVAP with larger power, the SEP can be divided into more free frequency band shares due to larger contribution, thereby being beneficial to improving the SEP utility (such as data rate). However, typically the maximum transmit power per SEP will be limited. If the power for charging is larger, the power for SEP to transmit its data will be smaller, which is not favorable for improving the SEP utility. Therefore, there is a need to find a power division that maximizes its utility. The relation is suitable for modeling as a non-cooperative game model, a group of SEPs providing charging service for the same SVAP are participants of the game model, and each participant performs self-transmitting power division by taking the maximized self-effect as a target.

The maximum transmit power of each SVAP is also fixed, and it is also necessary to find a power division that maximizes its utility, i.e., how much to forward data for an edge wireless terminal associated therewith, and how much to transmit its own data. Therefore, the SBS collects a portion of the spectrum resources of an associated edge wireless terminal of each of the SVAPs it elects as a reward for the set of SVAPs to forward data contributions, and determines the allocation of a reward share based on the contribution of each SVAP to its associated edge wireless terminal to forward data.

Therefore, if a certain SVAP forwards a large amount of data to its associated edge wireless terminal, in theory, in addition to the free spectrum resources contributed by its associated edge wireless terminal, some or all of the free spectrum resources contributed by other SVAP-associated edge wireless terminals will be obtained. However, in practice, less power is left for transmitting data by itself, since more transmit power must be paid for in order to obtain more spectrum resources. It may not be necessary for the excess spectrum resources obtained to bring more utility to itself. Based on a non-cooperative gaming model like the one described above, each SVAP can find a power partition that maximizes its utility.

After each SEP and SVAP feeds back the power division result thereof to SBS, SBS will evaluate the use efficiency of unit frequency band resource of edge wireless terminal, i.e. whether the division of the above three parts is reasonable or not. In the initial state, the SBS does not know what spectrum resource division ratio will maximize the benefit of the unit spectrum resource, and then sets an initial spectrum resource division ratio and sends it to each SEP and SVAP. Therefore, the aforementioned power division fed back by each SEP and SVAP is a result based on the initial unit spectrum resource division ratio. And after the SBS obtains the feedback of the power division, evaluating whether the unit frequency spectrum resource division can be further optimized. And if the optimization can be further carried out, updating the unit spectrum resource division result. Although the change of the power division result of each SEP and SVAP is not obvious based on the updated unit spectrum resource division result, the utility of each SEP and SVAP may change more obviously. If either the SEP or the SVAP is not satisfied with its own utility, it may be considered to not participate any more, which may result in an insufficient number of participants and an inability to build a collaboration framework. Therefore, SBS must consider avoiding this as much as possible. This may take into account the utility of each SEP and SVAP when the SBS constructs the utility function. The above relationship constitutes a typical Stackelberg gaming relationship. Here SBS is the leader in the Stackelberg game theory and each SEP and SVAP is the follower. After the leader makes a decision, the follower forms own decision based on the decision of the leader, and finally forms a Nash equilibrium state of the Stackelberg game, namely, each party does not change the decision.

Before elaborating the algorithmic description of the method of the invention, the necessary calculation formulas and parameters are illustrated as follows: when a receiving node j operates in an energy receiving mode, the energy it receives from a transmitting node i can be estimated by equation (1):

in the formula (1), the first and second groups,

and

respectively representing the transmitting power of a transmitting node i and the receiving power of a receiving node j; eta_i,jIs the energy conversion efficiency factor between the sending node i and the receiving node j, in the present invention, an empirical value of 0.56 is used; ag_i,jRepresenting the channel gain between the transmitting node i and the receiving node j, can be estimated by equation (2):

in the formula (2), G_t、G_rGains of a transmitting antenna and a receiving antenna respectively; h is_tAnd h_rRespectively representing the heights of the transmitting antenna and the receiving antenna from the ground; λ is the wavelength of the carrier signal; d_i,jIs the linear distance between the sender i and receiver j(ii) a L is a system loss factor independent of propagation; d_crossoverIs a compound defined in the document [2 ]]The cross distance of (2).

Information decoding rate b from transmitting node i to receiving node j when receiving node j is operating in information decoding mode_i,jAccording to shannon's theorem, can be estimated by equation (3):

in formula (3), W and σ²Representing the transmission bandwidth and the channel noise power around wireless node j, respectively. F_jThe interference power perceived by the wireless node j mainly consists of interference generated when the node using the same frequency band in a cell adjacent to the node j transmits data, and can be estimated by a formula (4):

in equation (4), Ag_k,jIs the channel gain on the link from interfering node k to interfered node j, which can be estimated using equation (2); p is a radical of_kIs the transmit power of interfering node k; i is_jIs the set of interference source nodes for node j.

For any SEP (e.g. e), if the maximum transmit power is used

Indicating charging power for charging SVAP v

Indicating the transmission power for transmitting its own data to SBS s

Expressed, the following relationship is required to be satisfied between the three components:

when the total transmission power of SEP e is

The energy stored in the battery is E_eWhen it is in continuous operation (i.e. during continuous operation)

) Can be estimated by equation (6):

in that

Meanwhile, SEP e is guaranteed not to move, and the net profit can be estimated by equation (7), which means that SEP e can transmit the total data amount of its own data to SBS using the acquired free spectrum resources.

In the formula (7), phi_eThe non-private contributing node, which is considered as a fictitious node, is usually set to a constant (e.g. 10-⁶) The purpose is to prevent game participants from communicating cheating in each round of game;

represents the contribution ratio of SEP e to the charging of its associated SVAP; r_vIs the set of charging nodes that charge SVAP v. The best response strategy for SEP e is to solve the following optimization problem (8). In theory, SEP e needs to find the optimal transmit power

Charging SVAP v to maximize its own profit can be obtained by solving (8). In fact, it is not straightforward to solve the transcendental equationThen, the result is obtained through formula transformation, so an algorithm based on a binary search idea is set forth in the following to solve the problem.

For any SVAP (e.g., v), it follows | R_vThe energy received by | SEPs can be estimated by the following equation, where | R_vI represents the set R_vThe number of middle members.

For any SVAP (e.g., v), if the maximum transmit power is used

Means for receiving data of its associated edge wireless terminal and forwarding the data to the forwarding power of the SBS s

Means for transmitting transmission power for transmitting data to SBSs

Expressed, the following relationship is required to be satisfied between the three components:

when the total transmission power of SVAP v is

And the energy stored in the battery and the received energy are respectively E_e、e_vWhen it is in continuous operation (i.e. during continuous operation)

) Can be prepared fromEstimating the formula:

in that

Meanwhile, the SVAP v needs to be guaranteed not to move, and the net benefit can be calculated by formula (12), which means that the SVAP v can send the total data volume of its own data to the SBS using the acquired free spectrum resources.

In the formula (12), | S_sI represents the number of SVAPs elected by the SBS; in the invention, without loss of generality, only one SVAP is considered to provide relay service for one edge wireless terminal, so | S_sI also represents the number of edge wireless terminals which can be coordinated by the SBS for frequency band resource scheduling; in the following description of the embodiments, | S_s| is set to 8; and setting phi_eFor the same purpose, parameter phi_vIs also set to prevent game participants from colluding in each round of the game, but is set to a different constant (e.g., 2). The best response strategy for SVAP v is to solve the following optimization problem (13). In theory, SVAP v needs to find the optimal transmission power

To provide data forwarding services for the edge wireless terminals associated therewith, thereby maximizing self-revenue, can be obtained by solving (13). In fact, since the transcendental equation is involved in the solution, the result cannot be obtained directly through formula transformation, so we set forth an algorithm based on the binary search idea to solve in the following.

For SBS, on the one hand, it wants to receive more edge wireless terminal data forwarded by SVAPs, and on the other hand, it needs to reasonably schedule a portion of the edge wireless terminal's spectrum resources to be used by SVAPs providing forwarding services and SEPs providing charging services to motivate and maintain their service motivation. Therefore, the present invention sets two excitation coefficients, i.e., the excitation coefficient ψ of SVAP_vAnd the excitation coefficient psi of the SEP_e. The values of the two excitation coefficients are both larger than 0 and smaller than 1. When the total amount of frequency band resources obtained by an edge wireless terminal from the SBS is W, the SBS expects xi_vW can be orchestrated by it to reward the forwarding contribution of SVAPs, and ξ_eW can be orchestrated by it to reward SEPs that provide charging services to the SVAP that forwards data for this edge wireless terminal. Frequency band resource division proportionality coefficient xi_eAnd xi_vAll values of (a) are more than 0 and less than 1, and simultaneously 1-xi_e-ξ_v>0。

The utilization of the edge wireless terminal resource block W of interest to SBS s can be estimated by equation (14), which means the total amount of data that SBS s wishes to receive from the edge wireless terminals. Here, if at

The larger the data amount in the period, the higher the utilization rate of the resource block W.

In equation (14), U represents an edge wireless terminal, and as mentioned above, the present invention only considers the case where one SVAP provides forwarding service for one edge wireless terminal, and thus each edge wireless terminal set (e.g., U)_v) Only 1 member of (a). Excitation coefficient psi of SVAP_vAnd the excitation coefficient psi of the SEP_eCan be estimated by equation (15):

in the formula (15), the first and second groups,

the total data volume of all SEPs sending own data to the SBS is represented;

representing the total amount of data that all SVAPs send their data to SBS.

And

can be estimated by equation (16):

in the formula (15)

And

are respectively

And

the theoretical maximum value. Let ξ e be 1, which can be estimated by equation (16-1)

A value of (d); xi (xi)_v1, can be estimated by equation (16-2)

The value of (c). In theory, the optimal response strategy for SBS s can be obtained by solving an optimization problem (17):

in fact, the optimization problem (17) can be transformed into a question about ξ_eAnd xi_vThe binary transcendental equation of (c). According to the document [3]In theory, a binary optimization problem can be approximately decomposed into two unary optimization problems to be solved. However, the decomposed two unary optimization problems still belong to transcendental equations and cannot be directly solved through algebraic operation. Based on the same thought, the invention adopts an algorithm based on the half-searching thought to solve. In order to simplify the following algorithm description based on the binary search concept, an abstract utility function μ (x) is uniformly adopted, and the actual utility function corresponds to the following.

The algorithmic description of the present invention is divided into two parts. The former section relates to the role construction of the incentive framework, including SBS election SVAPs, and election of a set of SEPs and association of a set of edge wireless terminals by each of the elected SVAPs. The latter part is a forming process of an excitation mechanism, and comprises a non-cooperative game process inside each group of SEPs, a non-cooperative game process among SVAPs, and a Stackelberg game process formed by interaction of SBS with different roles such as SVAPs and SEPs, and finally a spectrum resource matching scheme with maximized SEP and SVAP utilities is obtained.

The algorithmic process of the present invention has been described in detail above and will not be described in detail here.

The present invention will be specifically explained below with reference to the accompanying drawings in an embodiment of the present invention.

The radius of the micro-cellular network is 300 meters, and the micro-cellular network comprises an SBS positioned in the center of an area and a plurality of nodes which are randomly distributed, wherein the number of the nodes is 300. In this embodiment, we set 8 reference points to assist the base station in electing SVAPs, where the radius of 8 reference points is R_thAre uniformly distributed on the circle of (1), wherein R_thThe value of (d) was taken to be 60 meters. Per SVAP election6 SEPs to provide charging service for them. The selection conditions for SEPs and SVAPs are described in detail in the summary of the invention. The wireless terminals associated with each SVAP are located near the edge of the microcellular network because they are more in need of the assistance of SVAPs. Without loss of generality, we only focus on the case where each SVAP serves only one network edge wireless terminal. The simulation parameters are shown in table 1.

TABLE 1 simulation parameters

The results shown in fig. 2 to 5 were obtained using an OMNeT + +4.6 network simulator. As can be seen from fig. 2, the spectrum efficiency of the edge wireless terminal with SVAP assistance is significantly better than that of the case without SVAP assistance, which indicates that SVAP setting is necessary. In addition, we also observe that the node density in the microcellular network has randomness on the influence of the spectrum efficiency of the SBS, as the node density increases, the selection of the SVAP tends to be more close to the wireless terminal of the reference point, but closer to the reference point does not mean closer to the SBS, so the result has randomness, and as the node density increases, the edge wireless terminals associated with the SVAP will change randomly, and therefore the spectrum efficiency will also change accordingly.

As can be seen from fig. 3, as the channel noise power increases, there is a trend that the average spectrum efficiency of the edge wireless terminal when the SVAP assists in transferring data and the average spectrum efficiency of the edge wireless terminal when the SVAP does not assist in transferring data both decrease, however, the increase rate of the average spectrum efficiency increases, which shows that the role of the SVAP in improving the spectrum efficiency of the unit resource block is more and more obvious as the channel noise power increases. When the channel noise power is relatively small, the average spectrum efficiency of the edge wireless terminal when the SVAP helps to transfer the data is even higher than that of the edge wireless terminal when the SVAP helps to transfer the data, because: when the channel noise power is low, the edge wireless terminal directly sends data to the SBS, so that the edge wireless terminal can meet the own traffic demand and has higher spectrum efficiency, and when the SVAP assists, a part of own spectrum resources needs to be divided as excitation for free use by the SVAP and the SEP, but a larger benefit is not obtained. With the continuous increase of the channel noise power, the spectrum efficiency of the edge wireless terminal without the SVAP-assisted transit data and the spectrum efficiency of the edge wireless terminal with the SVAP-assisted transit data both continue to decrease until infinitely close to 0, but the situation of the excessive channel noise power is not common.

As can be seen from fig. 4, the influence of the node density in the microcellular network on the average spectral efficiency of all SEPs and the average spectral efficiency of all SVAPs has randomness, and as the node density increases, the selection of SVAP tends to be more toward the nodes closer to the reference point, thereby causing the preselection of SEP and SVAP to change, but closer to the reference point or closer to SVAP does not mean closer to SBS. Furthermore, variations in the locations of the SEP and SVAP mean variations in the resulting steady power, which are random as the locations.

From fig. 5, it can be seen that as the channel noise power increases, the average spectrum efficiency of SEP and SVAP using free spectrum resources tends to decrease, because: the increase in channel noise power tends to decrease the signal-to-noise ratio, resulting in a decrease in the overall spectral efficiency. In addition, it is observed that SEPs and SVAPs help the edge wireless node to improve efficiency, and also achieve considerable benefits, and therefore, there is a motivation that SEPs and SVAPs help the edge wireless node to improve efficiency.

Reference to the literature

[1]Cisco Visual Networking Index:Forecast and Methodology,2016–2021,CISCO White Paper,2017.

[2]WB Heinzelman,Application-specific protocol architectures for wireless networks,Ph.D.Thesis,Massachusetts Institute of Technology(2000).

[3]S.Boyd,L.Vandenberghe,Convex Optimization,Cambridge,U.K.:Cambridge Univ.Press,2004.

Claims (9)

1. A method for improving the spectral efficiency of a wireless terminal at the edge of a cellular weak coverage area, comprising the steps of:

(1) planning a plurality of reference points in a cellular weak coverage area, and selecting a wireless terminal with energy storage advantages and geographical position advantages as a virtual access point (SVAP) for each reference point according to a preset standard;

(2) each SVAP is responsible for recruiting wireless terminals with energy storage advantages and geographic position advantages to serve as the role of SEPs of small cell energy providers, each SVAP independently selects a group of SEPs to provide charging service for the SVAP, and the group of SEPs also obtains the frequency spectrum of a free access base station as the return for providing the charging service;

(3) each SVAP is associated with a group of edge wireless terminals so as to provide relay forwarding service for the edge wireless terminals conveniently, and a part of the frequency spectrum resources of the edge wireless terminals is exchanged according to the contribution of forwarding data and used as a free access base station;

(4) the SBS gives a division ratio of initial spectrum resources, starts a game process, and divides unit spectrum resources obtained by each edge wireless terminal into three parts, wherein one part is used for sending data of the edge wireless terminal, and the other two parts are respectively used for rewarding SVAP helping the edge wireless terminal to forward the data and a group of SEPs for charging the SVAP; SBS is small cell base station;

(5) each SEP in each group of SEPs determines the power for providing charging through a non-cooperative game process; after the charging power of all SEPs is obtained, each SVAP determines the power of the SVAP for forwarding data for the edge wireless terminal through a non-cooperative game process, and the SBS updates the division ratio of the frequency spectrum resources based on the feedback results of the SVAPs and the SEPs;

(6) and (5) repeating the step (5) until the variation value of the partition ratio of the unit spectrum resources is smaller than a preset threshold value, obtaining a Nash equilibrium point of the Stackelberg game, and obtaining a spectrum resource matching scheme with the SEP and SVAP maximized utility.

2. The method according to claim 1, wherein the algorithm procedure of SBS voting SVAPs in step (1) is as follows:

1) the election operations performed by SBS s:

step 1.0: initializing set S_sIf the SVAPs are empty, the selected SVAPs are stored, and then the step 1.1 is carried out;

step 1.1: determining the number of SVAPs needing to be elected, determining the positions of reference points according to the number and the distribution condition of the areas needing to be reinforced, wherein the number of the reference points is the same as the number of the SVAPs, and then entering step 1.2;

step 1.2: if the reference point is not traversed, broadcasting an SVAP election packet with the maximum transmission power aiming at an unprocessed reference point, then setting a time interval to wait for the return of a response packet, and entering the step 1.3; otherwise, the operation executed by the SBS s is ended;

step 1.3: if the time interval delta₁If the consumption is not finished, giving up the wireless terminal meeting the 5 SVAP election conditions in the previous round if the received response packet meets the 5 SVAP election conditions, reserving the wireless terminal meeting the 5 SVAP election conditions at present, and continuously waiting for the next response packet; otherwise, sending SVAP confirmation packet to the reserved wireless terminal, determining the SVAP confirmation packet as the SVAP of the currently investigated reference point, and storing the SVAP confirmation packet in the set S_sThen returning to the step 1.2;

2) and the wireless terminal participating in SVAP election executes the following operations:

step 2.0: if an SVAP election packet aiming at a certain reference point is received and the first 4 SVAP election conditions are met, sending a response packet to SBS s, and entering the step 2.1; otherwise, quitting;

step 2.1: if receiving the SVAP confirmation packet from the SBS, marking the identity of the SVAP as the SVAP of a certain reference point, and then finishing the operation; otherwise, waiting for SVAP election packages aiming at other reference points or exiting in time-out.

3. The method of claim 2, wherein in step (2), the algorithm procedure of the SVAP electing SEP is as follows:

1) election operations performed by SVAP v: one SVAP in the SVAPs is represented by SVAP v;

step 3.0: initializing the set R_vIf empty, to store the selected SEPs, then go to step 3.1;

step 3.1 if set R_vIf the number of members does not reach the preset value, broadcasting an SEP election packet with the maximum transmitting power, then setting a time interval to wait for the response packet to return, and then entering step 3.2; otherwise, the election operation executed by the SVAP v is finished;

step 3.2: if the time interval delta₂If the consumption is not finished, the received response packet is discarded for the wireless terminal meeting the 5 SEP election conditions in the previous round if the received response packet meets the 5 SEP election conditions, the wireless terminal meeting the 5 SEP election conditions at present is reserved, and the next response packet is continuously waited; otherwise, sending SEP confirmation packet to reserved wireless terminal and storing it in set R_vThen returning to the step 3.1;

2) and the wireless terminal participating in SEP election executes the following operations:

step 4.0: if an SEP election package of a certain SVAP is received and the first 4 SEP election conditions are met, sending a response package to the SVAP v, and entering the step 4.1; otherwise, quitting;

step 4.1: if an SEP confirmation packet from a certain SVAP is received, the SEP with the identity of the SEP as the SVAP v is marked, and then the operation is ended; otherwise, waiting for the SEP election package of the SVAP or exiting in time.

4. The method of claim 3, wherein in step (3), the algorithm for SVAP to associate a group of edge wireless terminals is as follows:

1) response association operations performed by SVAP v:

step 5.0: initializing set U_vEmpty, ready to store the associated edge wireless terminal, go to step 5.1;

step 5.1: if receiving the association request packet of a certain edge wireless terminal, feeding back an association response packet with the maximum transmission power, then setting a time interval to wait for the return of the association confirmation packet, and then entering step 5.2;

step 5.2: if the time interval delta₃If not, the sender is stored in the set U according to the received associated acknowledgement packet_vContinuing to wait for the next association confirmation packet; otherwise, the response association operation executed by the SVAP v is finished;

2) and executing the operation of the associated edge wireless terminal:

step 6.0: broadcasting an association request packet at maximum transmission power, then setting a time interval to wait for an association response packet to return, and then entering step 6.1;

step 6.1: if the time interval delta₄If not, reserving the sender of the association response packet with the strongest current received signal aiming at each feedback association response packet, and continuously waiting for the next association response packet; otherwise, sending an association confirmation packet to the reserved SVAP, and ending the association operation.

5. The method for improving the spectrum efficiency of the wireless terminal at the edge of the cellular weak coverage area according to claim 4, wherein the construction of the excitation mechanism among the SBS, the SEP and the SVAP is completed based on a binary search concept algorithm [ BinarySearch (min, max) ], and the specific process is as follows: the SBS gives an initial spectrum resource allocation ratio and starts a game process, a power division ratio is determined inside each set of SEPs through a non-cooperative game process, power division ratios are determined among the SVAPs through a non-cooperative game process, and the SBS determines whether to perform a next round of Stackelberg game process according to feedback results of the SVAPs and the SEPs.

6. The method of claim 5, wherein the specific process of the algorithm BinarySearch (min, max) is as follows:

step 7.0: setting two variables max and min which are respectively used for storing upper and lower boundary values of a reduced half search range, using the initial max and min of the value specified by the algorithm caller, and then entering step 7.1;

step 7.1: judging whether the absolute value of the difference value of max and min is larger than a threshold epsilon, if so, storing the value of (max + min)/2 by using a variable p, and entering the step 7.2; otherwise, the algorithm is ended, wherein epsilon is 0.1% of the difference between the upper and lower boundary values of the initial search range;

step 7.2: judging whether the utility function value with p-epsilon as an independent variable is larger than the utility function value with p as an independent variable; if yes, assigning the value of p to max; otherwise, assigning the value of p to min; then returns to step 7.1 where the utility function is according to the formula

And (6) estimating.

7. The method for improving the spectrum efficiency of the edge wireless terminal in the cellular weak coverage area according to claim 6, wherein in the step (4), the SBS initiates the Stackelberg game process to specifically operate as follows:

step 8.0: xi is reduced_eAnd xi_vAre all initialized to 0.2 and proceed to step 8.1 where ξ_eA division ratio, ξ, representing the spectral resource W_eThe divided partial resources are used for exciting a group of SEPs, the group of SEPs provides charging service for a certain SVAP, and the SVAP provides data forwarding service for an owner of the spectrum resource W; xi_vA division ratio, ξ, representing the spectral resource W_vThe divided resource can be used to reward the SVAP providing data forwarding service for the owner of the spectrum resource W in theory, but actually, the SBS is uniformly managed to stimulate the SVAP to select a group of SVAPs, i.e. forming a competition relationship, and the more contributor will obtain more being xi_vFree usage rights of the divided partial resources;

step 8.1: a variable C_eAnd C_vInitializing the SEPs and the SVAPs to be 0 for respectively storing the quantity of the SEPs and the SVAPs, and entering the step 8.2;

step 8.2: for each SEP, initializing the power to charge its associated SVAP to P_e ^max(ii)/2, initializing its power to transmit its own data to SBS s

Accumulating the number of SEPs and storing in C_eIn step 8.3, where P_e ^maxThe maximum transmitting power of SEP e is 200 milliwatts;

charging power for charging SVAP v; SEP e is the e-th SEP;

step 8.3: initializing the power P of forwarding data to the associated edge wireless terminal for each SVAP_v ^max(ii)/2, initializing its power to transmit its own data to SBS s

Accumulating the number of SVAP and storing in C_vIn step 8.4, where P_v ^maxThe maximum transmitting power of the SVAP v is 200 milliwatts;

a forwarding power for receiving data of the edge wireless terminal associated with SVAP v and forwarding the data to SBS s;

step 8.4: according to the formula

Calculating a utility function value, storing the utility function value in a variable mu expression, and entering the step 8.5; psi_eThe excitation coefficient of SEP; psi_vThe excitation coefficient of the SVAP is; w is the transmission bandwidth; t is_v ^maxWhen the total transmitting power of SVAP v is P_v ^maxAnd the energy stored in the battery and the received energy are respectively E_e、e_vThe continuous working time of the SVAP v; sigma²Is the channel noise power; f_sInterference power perceived for wireless node s; p_u ^maxIs the maximum transmit power of the wireless terminal; g_u,sChannel gain on the link from interfering node u to interfered node s; g_v,sFrom interfering node u to interfered node sChannel gain on the link of (1);

step 8.5: initializing variables

Is C_v/C_eInitialization of

Is 1/C_vThe process proceeds to step 8.6, where,

representing the fraction of SEP e contribution to charging SVAP v,

representing the contribution ratio of the SVAP v to the SBS s for the associated edge wireless terminal to forward data;

step 8.6: SBS s broadcast ξ to all SEPs and SVAPs within the served small cell area_e，ξ_v，

Four parameter messages.

8. The method of claim 7, wherein the step (5) comprises the following steps:

(A) non-cooperative gaming processes within each set of SEPs:

1) operation performed by the wireless terminal e serving as the SEP:

step 9.0: if receiving the message containing xi sent from SBS s_eAnd

in the message (2), the value of min is assigned to 0, and the value of max is assigned to P_e ^maxEntering step 9.1;

step 9.1: invoke the algorithm BinarySearch (min, max) to obtain power

Will be provided with

Assigned a value of P_e ^maxAnd

the difference between them

And

sending the value of (c) to SBS s, and entering step 9.2;

the transmission power for SEP e to send its own data to SBS s;

step 9.2: if the end packet sent by the SBS s is not received, the following operations are repeatedly executed: if received from SBS s transmission

The algorithm BinarySearch (min, max) is invoked to obtain power

Will be provided with

Is updated to P_e ^maxAnd

the difference between them

And

sends the value of (d) to SBS s and then returns to step 9.2; otherwise, go to step 9.3;

step 9.3: if an end packet is received from the SBS s, the SEP e will be

And

the value of (c) is sent to its associated SVAP v and the algorithm is ended; otherwise, returning to the step 9.2;

2) the response operations performed by SBS s for each SEP:

step 10.0: if receiving an inclusion from any SEP

And

if the received value is different from the corresponding value of the SEP e, the newly received value is used for replacing the original value; otherwise, accumulating the number of SEPs with unchanged power division values, and entering the step 10.1;

step 10.1: if at least one SEP in the received messages from SEPs updates the power division value, then according to the formula

Computing

And broadcast an include

The update package of (1) is sent to all SEPs, and the step 10.2 is entered; t is_e ^maxWhen the total transmission power of SEP e is P_e ^maxThe battery stores energy as E_eThe continuous working time of SEP e;

the contribution ratio when SEP e charges its associated SVAP; eta_e,vIs an energy conversion efficiency factor; phi is a_eIs a fictitious non-private contribution node; r_vA set of charging nodes to charge the SVAP v; g_e,sIs the channel gain on the link from node e to node s; g_e,vIs the channel gain on the link from node e to node v; e' denotes all the residues belonging to the set R_vA node of (2); eta_e',vRepresenting an energy conversion efficiency factor from node e' to node v;

represents the charging power from node e' to the charge of node v;

step 10.2: if the number of the SEPs with unchanged power division values reaches the maximum value of the system, broadcasting an end packet to all SEPs; otherwise, returning to the step 10.0;

(B) non-cooperative gaming process between SVAPs:

1) operation performed by the wireless terminal v acting as SVAP:

step 11.0: if receiving the signal containing xi sent by SBS s_vAnd

and all related SEPs from the message

And

in the message (2), the value of min is assigned to 0, and the value of max is assigned to P_v ^maxEntering step 11.1;

step 11.1: invoke the algorithm BinarySearch (min, max) to obtain power

Will be provided with

Assigned a value of P_v ^maxAnd

the difference between them

And

sending the value of (c) to SBS s, and entering step 11.2;

step 11.2: if the end packet sent by the SBS s is not received, the following operations are repeatedly executed: if received from SBS s transmission

The algorithm BinarySearch (min, max) is invoked to obtain power

Will be provided with

Is updated to P_v ^maxAnd

the difference between them

And

sends the value of (d) to SBS s and then returns to step 11.2; otherwise, go to step 11.3;

step 11.3: if receiving the end packet sent by SBS s, ending the algorithm; otherwise, returning to the step 11.2;

2) the response operations performed by SBS s for each SVAP:

step 12.0: if receiving an inclusion from any of the SVAPs

And

if the received value is different from the original value, the newly received value is used to replace the original value; otherwise, accumulating the number of SVAPs with unchanged power division values, and entering step 12.1;

step 12.1: if at least one SVAP updates the power division value in the received messages from the SVAPs, then the power division value is updated according to the formula

Computing

And broadcast an include

The update package of (2) is sent to all SVAPs, and step 12.2 is entered;

charging power from node v to charge node s; v, all belong to the set S_sA node of (2); g_v、,sIs the channel gain on the link from node v to node s;

step 12.2: if the number of the SVAPs with unchanged power division values reaches the maximum value of the system, broadcasting an end packet to all the SVAPs; otherwise, return to step 12.0.

9. The method of claim 8, wherein in step (5), the SBS performs a Stackelberg game process, and the specific steps are as follows:

(C) the SBS updates the division ratio of the frequency spectrum resources based on the feedback results of the SVAPs and the SEPs:

step 13.0: if the end packet has been sent to all the SVAPs, assigning the value of min to 0, assigning the value of max to 1, and entering step 13.1;

step 13.1: let xi_eKeeping the current value unchanged, call the algorithm BinarySearch (min, max) to get the new ξ_vAnd entering step 13.2;

step 13.2: let xi_vKeeping the value obtained in step 13.1 unchanged, calling the algorithm BinarySearch (min, max) to obtain new xi_eAnd entering step 13.3;

step 13.3: with the current newest xi_eAnd xi_vIs an input parameter, by formula

Calculating a utility function value, storing the utility function value in a variable mu' expression, and entering a step 13.4;

step 13.4: judging whether the mu 'is more than mu or not, if so, assigning the mu as mu', and returning to the step 8.5; otherwise, the algorithm is ended, wherein the initial value of μ is obtained by step 8.4.

Images