CN113163495A

CN113163495A - Bilateral dynamic multi-demand ascending auction spectrum allocation method based on cloud cluster purchase platform

Info

Publication number: CN113163495A
Application number: CN202110081585.4A
Authority: CN
Inventors: 李志先; 邵鸿翔; 李蒙; 丁思淼; 陈博; 陈耀恒; 王华鑫
Original assignee: Luoyang Institute of Science and Technology
Current assignee: Luoyang Institute of Science and Technology
Priority date: 2021-01-21
Filing date: 2021-01-21
Publication date: 2021-07-23
Anticipated expiration: 2041-01-21
Also published as: CN113163495B

Abstract

According to the bilateral dynamic multi-demand ascending auction spectrum allocation method based on the cloud group purchase platform, a seller reports a resource capable of being shot; the platform divides resources into resource blocks and issues resource block information; the buyer can quote price and report the buyer position information according to the self requirement and the information of the frequency spectrum resource block of the platform; the platform divides independent sets according to the position information of the buyers to form group buying groups and carries out auction; and finally, comparing the two sides of the result to finish the transaction. The buyer obtains the spectrum usage right for the period and pays. The method guarantees the benefits of both buyers and sellers, gives consideration to the fairness of distribution results, avoids the possibility that resources are monopolized by a few main bodies, distributes all spectrum resources comprehensively and reasonably, and improves the spectrum utilization rate.

Description

Bilateral dynamic multi-demand ascending auction spectrum allocation method based on cloud cluster purchase platform

Technical Field

The invention relates to the technical field of spectrum resource allocation, in particular to a bilateral dynamic multi-demand ascending auction spectrum allocation method based on a cloud shopping platform.

Background

For the fifth generation (5G) and future wireless communications, network size and traffic volume are expanding to support ubiquitous communications of real-time mass access. The explosive growth of wireless traffic and the increasing demand for wireless spectrum resources have made the "spectrum deficit" problem increasingly prominent due to scarcity of wireless spectrum resources. Therefore, an innovative spectrum management mechanism is needed to break through the structural shortage problem of spectrum resources caused by the traditional static spectrum allocation mode, and improve the spectrum utilization rate. Dynamic spectrum sharing is considered one of the most directly effective means to alleviate the spectrum resource shortage problem. Some Licensed bands are not fully utilized, leading to the idea of granting Licensed Shared Access (LSA).

In the conventional cognitive radio sharing concept, secondary users, such as secondary mobile network operators, cannot guarantee access to the authorized spectrum. In LSA, the duration and conditions of the shared spectrum are precisely defined in advance by regulatory bodies through licenses. Deploying an LSA system requires the introduction of two new architectural blocks, including an LSA database (information about the LSA frequency band, such as sharing conditions, bandwidth and duration, etc.) and an LSA controller (for controlling access to the LSA bandwidth). The invention particularly considers the reusability of frequency spectrum (base stations without interference relationship can use the same frequency spectrum band at the same time), and realizes that a plurality of Base Stations (BSs) of different operators compete for LSA frequency spectrum in a limited time of a specific geographic area by a virtual ascending auction method and cloud platform operation. By way of the concept of a typical spectrum auction, the terms bidder, player and Base station are used herein to refer to buyers in an auction context. In the invention, the platform is used as a mediator to establish a spectrum pool for all idle spectrum resources to be sold by the seller to perform integration processing.

At present, the frequency spectrum auction of the LSA mainly adopts an all-or-nothing mode, namely an all-or-nothing scheme, namely all available LSA frequency spectrums are regarded as an indivisible block and can be distributed to a buyer body only, and other buyers participating in the auction are not acquired, so that the flexibility and the fairness are lacked. Second, they are one-time auctions. The one-time auction mechanism consists of a single round of auctions so bidders have only one opportunity to submit bids to the auctioneer. The one-time auction mechanism also lacks flexibility and fairness, and the situation that the buyer can not obtain the one-time auction mechanism exists.

Disclosure of Invention

In order to solve the technical problems, the invention provides a cloud cluster purchase platform-based bilateral dynamic multi-demand ascending auction spectrum allocation method, which guarantees the benefits of buyers and sellers, gives consideration to the fairness of allocation results, avoids the possibility that resources are monopolized by a few main bodies, allocates all spectrum resources comprehensively and reasonably and improves the spectrum utilization rate.

In order to realize the technical purpose, the adopted technical scheme is as follows: the bilateral dynamic multi-demand ascending auction spectrum allocation method based on the cloud group purchase platform comprises the following steps:

s1, the frequency spectrum auction system comprises three parties: sellers, buyers and cloud buying a spectrum platform; the cloud cluster frequency spectrum purchasing platform receives all frequency spectrum resources reported by sellers and corresponding frequency spectrum resource quotations, divides all frequency spectrum resources into k frequency spectrum resource blocks, and receives respective bids B of the buyers on the k frequency spectrum resource blocks_iBid price B_iThe bidding values are arranged from high to low, the cloud group purchase frequency spectrum platform divides all buyers into h group purchase groups, and obtains group purchase group bidding vectors of each group purchase group for k frequency spectrum resource blocks

S2, the cloud cluster frequency spectrum purchasing platform auctions k frequency spectrum resource blocks, and performs frequency spectrum allocation by using an ascending frequency spectrum auction method;

s3, calculating the actual payment value of each buyer for the distributed spectrum resource block according to the respective bid of the buyer in the same group purchase group distributed to the spectrum resource block and the respective bid of the buyer in other group purchase groups not distributed to the spectrum resource block;

s31, confirming that a group purchase group obtains the frequency a of the allocated spectrum resource blocks;

s32, bidding vectors for the rest group buying groups without obtaining the spectrum resource block

Sorting, arranging the first k values from low to high to form a competition vector

S33, obtaining the actual payment value of the buyer in the group purchase group of the b-th distributed spectrum resource block as a competition price vector

Minus the b-th value of the remaining buyers of the same group, b being 1, 2, … …, a;

s34, repeating the steps S31-S33, and calculating the actual payment value of the spectrum resource blocks distributed by all buyers in all group purchase groups;

and S4, according to the actual payment value of each buyer to the same spectrum resource block obtained in the step S3, the total price of all spectrum resource blocks which belong to the same spectrum resource and are calculated by the cloud cluster frequency spectrum purchasing platform is compared with the frequency spectrum resource quotation of the seller of the spectrum resource, if the total price is higher than the frequency spectrum resource quotation, the transaction is carried out, and if the total price is lower than the frequency spectrum resource quotation, the streaming is carried out.

The group buying group is divided according to the geographical location information and the maximum transmitting power of all the buyers.

The specific implementation method of step S2 is:

s2.1, uniformly calling k spectrum resource blocks by the cloud cluster spectrum purchasing platform from p to 0;

s2.2, bidding vectors for group buying groups of each group buying group higher than the current round p value

Inner postCounting the paid number of the resource blocks with frequency spectrum to obtain the number of the resources which can be shot in each group purchase group in turn, and removing one group purchase group G in turn_hCalculating the total number of the clapable resources of the rest group purchase group and the total number of the clapable resources of all the group purchase groups;

s2.3, if the total number of the resources which can be shot of all the rest group purchase groups is more than or equal to the k value, entering the next round of auction price calling, taking the value delta added when the value p of the round is used as the value p of the next round, continuing to call the price, and repeating the step S2.2;

s2.4, if the total number of the resources which can be shot of one remaining group purchase group is less than the k value and the total number of the resources which can be shot of all the group purchase groups is more than the k value, the cloud group purchase spectrum platform divides one spectrum resource block to provide the removed G for the cloud group purchase spectrum platform_h' taking the p value of the current round as the p value of the next round, subtracting 1 from the k value, continuing to call the price, and repeating the step S2.2;

and S2.5, if the total number of the resources which can be bought by the group purchase groups is equal to the initial number of the spectrum resource blocks, the cloud group purchase spectrum platform allocates the spectrum resource blocks according to the number of the resources which can be bought by the group purchase groups according to the spectrum resource blocks, so that each group purchase group is guaranteed to obtain at least one spectrum resource block, and the pricing is finished.

The invention has the beneficial effects that:

1. by dividing a large segment of frequency spectrum into small resource blocks for flexible transaction, resource waste is avoided. And improves the performance of the auction by a mechanism designed to increase the fairness of the allocation (as the spectrum is not necessarily allocated to one or only one group). The VCG-like payment rules are used to maintain their incentive properties (true bids). The privacy of the winning bidders is protected and the valuation is not revealed using the ascending auction. In addition, they are more transparent because each bidder can see the evolution of the auction. The auction mechanism provided by the invention can be proved to meet the economic characteristics in the aspects of authenticity, personal rationality after the fact, cost budget balance, economic efficiency and the like.

2. The invention provides a new idea for spectrum allocation based on convergence, effectiveness, transparency and privacy of bilateral dynamic multi-demand ascending auction spectrum allocation of a cloud buying platform.

Drawings

FIG. 1 is a schematic diagram of the system of the present invention;

FIG. 2 is a diagram of the actual bid versus the number of spectral resource block demands of the present invention;

FIG. 3 is a graph showing actual bid for different services according to the present invention;

fig. 4 is a schematic diagram of a group purchase group composed by buyers according to the present invention.

Detailed Description

The bilateral dynamic multi-demand ascending auction spectrum allocation method based on the cloud group purchase platform comprises the following steps: s1, the frequency spectrum auction system comprises three parties: sellers, buyers and cloud buying a spectrum platform; the cloud cluster frequency spectrum purchasing platform receives all frequency spectrum resources sent by sellers and corresponding frequency spectrum resource quotations, divides all frequency spectrum resources into k frequency spectrum resource blocks, and receives respective bids B of the buyers on the k frequency spectrum resource blocks_iBid price B_iThe bid values in the group are arranged from high to low, all buyers are divided into h group purchase groups, and group purchase group bid vectors of each group purchase group for k frequency spectrum resource blocks are obtained

The entire spectrum auction mechanism involves three parties, authorized spectrum holder-seller (seler), secondary spectrum user-buyer (buyer), spectrum auction cloud group buying spectrum platform-arbitrator. Secondary spectrum users, typically secondary base stations, have spectrum requirements and are the buyers of spectrum transactions. As a seller, an organization entity with authorized spectrum usage rights has a large amount of free spectrum resources in a time period and has a circulation willingness. Due to the spatial multiplexing property of the spectrum resources, different base stations far enough away can use the same spectrum at the same time. And the cloud group purchase frequency spectrum platform finishes the whole processes of auction reporting, auction and payment.

As shown in fig. 1, on one hand, a seller main body reports a base price and spectrum resources available for short-term lease, and a cloud group purchase spectrum platform integrates and divides all spectrum resources reported by a seller and available for secondary circulation into a plurality of spectrum resource blocks to form resources in a spectrum circulation pool. On the other hand, the buyer base station reports information such as geographic position, power, quotation, spectrum (number of channels) requirements and the like, and the platform calculates all independent buyer integrated group buying groups and corresponding group buying group quotation according to the reported information. The group purchase group quotation is obtained by summing the quotations of the buyers in the group after the quotations are arranged from high to low.

The whole auction process is carried out by a cloud cluster frequency spectrum purchasing platform, and the cloud cluster frequency spectrum purchasing platform is used for efficiently matching the seller idle frequency spectrum with the buyer base station requirement by utilizing the advanced multi-resource-block auction method according to the pre-sold frequency spectrum, the frequency spectrum requirement, the geographic position, the corresponding bid and other information provided by the buyer and the seller. The leasing and auction of spectrum resources is performed in fixed time periods (e.g., once per t minutes/hour), each auction period comprising a plurality of time slots, involving 4 main phases, 1. buyer's bidding phase (to provide location, usage parameters, demand and bid information); 2. a platform virtual auction stage (completing resource matching); 3. a spectrum use stage (after the auction is finished, a buyer pays according to an actual auction result and temporarily has a spectrum use license until a use return platform of the spectrum is closed after a specified time); 4. and (4) ending the stage (the seller reports the spectrum leasing information of the next period, and the platform sorts and issues the number of the spectrum resource blocks of the next round to the buyer). The next round begins and the buyer's auction continues to participate in the auction for the spectrum resource. The whole platform design realizes effective dynamic circulation of frequency spectrum resources.

We assume that each buyer agent can request multiple channels. Let Seller be [ S ]₁,S₂,...,S_n]Indicating the vendors Seller who will provide the spectrum resource, n ═ Seller | indicates the cardinality (number of elements, number of vendors) of the vector set, and they provide the spectrum resource and the corresponding price as

And representing the corresponding quotation corresponding to each spectrum resource. The cloud group purchase spectrum platform establishes a spectrum pool available for auction according to spectrum resources provided by all sellers, and divides all the spectrum available for auction in the spectrum pool into k spectrum resource blocks.

Buyer layer ═ Ber₁,Ber₂,...,Ber_m]And m is | Buyer |, which represents the number of buyers. All buyer rootsAccording to the demand, the content of K frequency spectrum resource blocks is quoted according to increment₁,Β₂,...,Β_m]The urgency of the offer and the demand, the type of service and the time period associated, e.g. high definition video service at peak of service, offer beta_mEach unit channel in the vector has higher quotation and smaller quotation difference. Each buyer Beri has a different channel true value vector V for all accessible channels_i＝{v_i,1,v_i,2,..,v_i,k,..,

v

_i,K1, 2, … … m, reflecting the actual true value of Beri for different numbers of channels. Buyer keeping rationality, bidders tend to put the bid price below their own best estimate of the value of the common pre-beat spectrum (real value) because they are concerned about "cursing winners", i.e. losing too much. The bidding vector of each buyer is determined according to the actual business requirement, and the invention sets the constraint, the resource block bidding function V of each buyer i_i(.) is a non-decreasing convex function that will guarantee the authenticity of the buyer's bid in the auction. Since the present invention can guarantee the authenticity bid, the actual bid beta for buyer i_i＝[B_i,1,B_i,2,...,B_i,k,...,B_i,K]Will be related to its true value V_i＝{v_i,1,v_i,2,..,v_i,k,..,v_i,KThe same. We will consider the convexity of the estimates. We assume that the spectrum is divided into K identical spectrum resource blocks, and the buyer will quote different numbers of resource blocks according to his own needs. For each increased resource block demand, the corresponding price increment decreases as the number of resource blocks that have been acquired increases, resembling a logarithmic function, with a lower and lower rate of increase of the quote, as shown in fig. 2. First element B_i,1Representing a bid on the first resource block to be acquired, B_i,kThe quotation of 1 resource block is additionally obtained after the representative of the resource block acquisition system has obtained k-1 resource blocks. This amounts to a discretization of the convex valued function of the spectrum. The non-decreasing convex function setting also accords with the economic characteristic that the marginal effect of the actual demand is decreased, but in the actual operation, V_iOr BETA_iThe trend characteristic of the convex function can be determined according to actual serviceWith some variation. Different traffic may use different curve forms, as shown in fig. 3, general traffic uses curve 1, high definition spectrum uses curve 2 due to large bandwidth, and low traffic such as voice may use curve 3 due to less bandwidth.

The platform establishes a grouping group purchase mode according to the geographical position, the maximum transmitting power and the range information of the buyer, and improves the frequency spectrum use efficiency and reduces the user cost by utilizing the spatial reuse property of the frequency spectrum. The spectrum at a particular time within a particular geographic area may use the same bandwidth at the same time if two base stations do not interfere with each other. And the auction platform calculates the transmission range of the base station according to the power and the position reported by the buyer base station, and establishes an interference diagram and an independent auction buyer set. The independent buyers form a group buying group, the collective bidding vectors of the group buying group are formed by adding the elements of the bidding vectors of the group members, the h group buying group bidding vector is expressed as

Wherein

Represents the sum of the bids of all the buyers for the k-th spectrum resource block in the h-th group purchase group,

represents the sum of the bids of all the buyers to the 1 st spectrum resource block in the h group purchase group. Finding the largest independent set in the graph is an NP-Hard problem, and the invention does not relate to the relevant content of interference graph establishment. As shown in fig. 4, assuming there are 5 buyers and the corresponding interference relationship, the corresponding independent set form and the composition of the group purchase group can be obtained.

Authorized spectrum holders (sellers) report spectrum transfer usage information, relating to spectrum range and corresponding time omega (S)_x,S_y,T_x,T_y). The platform side establishes a time/frequency two-dimensional frequency spectrum pool according to the information reported by the seller, and the frequency spectrum in the pool is divided into k independent unequally-divided channels (the channel division can be divided according to the frequency spectrum category and the local service characteristics, and the channels are divided into packetsIncluding the number and extent of channels).

Secondary spectrum user (buyer) providing spectrum bid vector BETA_i＝[B_i,1,B_i,2,...,B_i,k,...,B_i,K]The platform judges the interference range of each base station and divides the maximum independent set of the base stations according to geographic data to form corresponding h group purchase groups G ═ G₁,G₂,...,G_h]Single group purchase group G_hThe bid vector is obtained by accumulating the corresponding elements of the bid vectors of the members in the independent set group and is expressed as

the specific implementation method of step S2 is:

Counting the prices of all the spectrum resource blocks in the group buying group, obtaining the number of the resources which can be shot in each group buying group in turn, and removing one group buying group G in turn_hCalculating the total number of the clapable resources of the rest group purchase group and the total number of the clapable resources of all the group purchase groups;

s2.4, if the total number of the resources which can be shot of one remaining group purchase group is less than the k value and the total number of the resources which can be shot of all the group purchase groups is more than the k value, the cloud group purchase spectrum platform divides one spectrum resource block to provide the removed G for the cloud group purchase spectrum platform_h' when the p value of the current round is added with the value delta as the p value of the next round, the k value is subtracted by 1, and the pricing is continuedAnd repeating the step S2.2;

The platform establishes a spectrum resource pool according to a pre-shot spectrum provided by a spectrum owner, and divides the resources in the pool into 5 channel resource blocks, wherein k is 5. If the p value of the first round is 0, the group buying group is three.

First group buying group G₁Consists of two buyers, the actual bid is { (12,10,5,3,1), (15,12,10,8,2) }, and the corresponding group buying group bid vector is

Second group buying group G₂Consists of two buyers, the actual bids are { (15,13,11,9,8), (13,12,11,8,6) }, and the corresponding group buying group bid vector is

Third group buying group G₃The method comprises the steps that a buyer is composed, the actual bid is { (26,20,10,9,1) }, and the corresponding group buying group bid vector is

In this case, p is 0, the number of spectrum resource blocks that can be paid by 3 group buying groups is 5, and the number of resource blocks that can be sold is greater than k (if p is 0, both are greater than 5) regardless of whether any group buying group is removed, so that there is no case where the existing number of resource blocks cannot be paid out by losing a certain group buying group. If the adding price amplitude is 1 each time, the auction is continued until the existing resource number cannot be played if a certain group buying group does not participate in the card.

The count above p is 5 and,

the count above p is 5 and,

counts above p are also 5, regardless of removal of G₁、G₂Or G₃If the sum of the counts of the other group purchase groups is equal to 10, the p-value is added to delta and is changed to 1, and then the steps are carried out

Counting higher than p value, counting at 5, 4, and removing G₁、G₂Or G₃Until the price is added.

S3, calculating the actual payment value of each buyer for the divided spectrum resource block according to the respective bid of the buyer in the same group purchase group allocated to the spectrum resource block and the respective bid of the buyer in other group purchase groups not allocated to the spectrum resource block;

s31, confirming that one group purchase group obtains the distributed frequency spectrum resource block number a;

For example: the platform establishes a spectrum resource pool according to a pre-shot spectrum provided by a spectrum owner, and divides the resource in the pool into 5 channel resource blocks, wherein k is 5 and delta is 1.

First group G₁Contains 2 persons, and the member bids are { (12,11,5,3,1), (15,12,10,8,2) }

Second group G₂Contains 2 persons, and the member bids are { (15,13,11,9,8), (13,12,11,8,6) }

Third group G₃Contains 1 person, the member bids { (26,20,10,9,1) }

According to the price quoted by each group member, the price quoted by the group buying of three independent sets (group buying groups) can be calculated, namely the sum of the corresponding items of each vector of the members (buyers) in the group. The group bid vectors for the three independent sets are:

with rising price, all things are

The corresponding requirement with the number smaller than p in the vector is eliminated, for example, p is 1, the third group cannot obtain 5 channels, and the corresponding D is then₃＝4。

And removing any one group purchase group from 1-14 times of calling, wherein the sum of the number D of the beat resources exceeding p of the other two group purchase groups is more than or equal to 5 of the number of the spectrum resources to be distributed (the situation that all resource blocks cannot be beat out without participation of the group purchase group does not exist). And no spectrum resource block is allocated, and p is equal to p + delta, and the price continues to be called in ascending order.

When p is 15, a first group of BETA₁ ^GIf 3 elements in the vector are not more than 15, only 2 beat resource numbers can be obtained, and at the moment, if the second group is not taken into calculation as Gh', and the total count of the first group and the third group exceeding 15 is 4, all 5 spectrum resource blocks cannot be beat, so that one spectrum resource block is allocated to the second group, and the number of the spectrum resource blocks to be allocated is changed into 4. When p is 20, the total count exceeding 20 in the first and third groups of bid vector elements is 3, 4 spectrum resource blocks to be allocated cannot be completed, one spectrum resource block is allocated again to the second group, the number of spectrum resource blocks to be allocated becomes 3, and the auction stops when p is 22 (the sum of demands is 5, and the total number of resource blocks provided by the seller).

The table above gives the increase of p per bid, the winning team and the payment for it. And (3) calculating the number D of the resources which can be shot and exceed the price each time, judging that the obtained channel number is less than D, and distributing the resource blocks to the frequency spectrum resource blocks, wherein when p is 22, the rest 3 resource blocks are left, D1+ D2+ D3 is 2+2+1 which is 5, which is exactly equal to the number of all the resource blocks which can be shot, the second group does not participate in the auction, all the resources are just shot, and the auction is ended. In addition, since the 15 th and 20 th G2 have been allocated with the spectrum resource block twice, and the D2 is not less than 2, no new spectrum resource block is allocated. At this time, the first group will get 2 resource blocks, the third group gets 1 resource block, and the whole allocation is completed.

Calculate the final payment:

when the value of the addition is increased to 15,

if the second group does not participate in the auction, the 5 channel resource blocks cannot be sold in their entirety. At this time, the second group locks its first spectrum resource block at p-15, and the payment fee of each bidder in the group purchase group is:

group buying group of first and third groups

And

the vectors are sorted, and the top 5 values are taken to form a second group of proper competitive price

The bids of two members of the second group are { (15,13,11,9,8), (13,12,11,8,6) }, and the second group is allocated to a frequency spectrum resource block for the first time and adopts

And other buyers' first value.

This actual payment paymoney for user 1 in the second group is:

this actual payment paymoney for user 2 in the second group is:

after the second group obtains one resource block, the number of resource blocks available for auction is shifted by 4.

At this point, the call continues and the p value continues to increase.

When the addition increases to p-20,

if the second group does not participate in the auction, the 4 spectrum resource blocks cannot be sold completely, at this time, the second group locks its second spectrum resource block at p-20, and the payment fee of each bidder in the group is the second payment of the second group, which is to pay attention to the use of the second bid at this time.

Group buying group of first and third groups

And

Two members of the second group bid { (15,13,11,9,8), (13,12,11,8,6) }respectively

This actual payment paymoney for buyer 1 in the second group is

This actual payment paymoney for buyer 2 in the second group is:

after the second group obtains one resource block, the number of resource blocks available for auction is changed by 3.

At this time, the price calling is continued, and the p value is increased once to perform judgment once.

When the call value p is 22,

and finishing the addition. At this time D₁＝2,D₂＝2,D₃The result of the last allocation is that the first group gets 2 spectrum resource blocks, the second group gets 2 spectrum resource blocks, and the third group or 1 spectrum resource block. Whereas the previous second group has already allocated 2 resource blocks and is therefore not participating in this allocation. Now calculate group buy group 1 get 2 resource blocks and group buy group 3 get 1 resource blockThen, the payment fee for each bidder in each group is:

a first group: two actual payments total

Group buying group of second and third group

And

the vectors are sorted, and the top 5 values are taken to form a proper amount of competitive prices of the first group

The two members of the first group bid { (12,11,5,3,1), (15,12,10,8,2) }respectively

The actual payment paymoney for this time for two spectrum resource blocks by buyer 1 in the first group is:

the actual payment paymoney for this time for two spectrum resource blocks by buyer 2 within the first group is:

third group: first actual payment

Group purchase group vector for first and second groups

Sorting, and taking the first 5 values to form a third group with proper competitive price

The third group has only one member, whose offer is { (26,20,10,9,1) }

This paymoney payment by buyer 1 in the third group is: ,

and adding the riveting points for 3 times to obtain a final auction result:

group 1 gets 2 spectral resource blocks, and 2 members pay 15 and 19, respectively.

Group 2 gets 2 spectral resource blocks, and 2 members pay 10 and 7, respectively.

Group 3 gets 1 spectrum resource block and 1 member pays 22.

At this time, allocation of 5 spectrum resource blocks is completed, and three groups of accumulated resource acquisition conditions and corresponding payment are calculated. And comparing the total price of all the spectrum resource blocks belonging to the same spectrum resource, which is calculated by the cloud cluster spectrum purchasing platform, with the frequency spectrum resource quotation of a seller of the spectrum resource, if the total price is higher than the frequency spectrum resource quotation, the transaction is carried out, and if the total price is lower than the frequency spectrum resource quotation, the streaming shot is carried out.

As can be seen from the verification example, the payment cost is less than the quotation, if the quotation is high, the resource is obtained, the payment is less, and the buyer is encouraged and ensured to carry out the true quotation. And, all the spectrum resource blocks can be tapped in the process.

In summary, the present invention relates to an auction mechanism for conducting short-time transaction on spectrum licenses under a limited attribute condition in an LSA scenario, which maximizes the benefits of buyers and sellers, improves the utilization rate of spectrum resources, and realizes effective circulation and redistribution of wireless spectrum resources by using the spectrum reuse characteristics in a group buying manner through the use and bid of potential spectrum buyers. The invention designs an auction mechanism considering the reusability of frequency spectrum, the authenticity, the fairness and the effectiveness of transaction. In the context of LSA spectrum, each spectrum buyer should be incentivized to indicate in good faith that they are willing to pay for good east and west, independent of other bidder behavior. The possibility that bidders generally try to operate the mechanism to maximize profits and damage profits of other players (including auctioneers) is avoided, the profits of buyers and sellers are guaranteed, fairness of distribution results is considered, and the possibility that resources are monopolized by a few subjects is avoided.

The LSA concept of the present invention has two main differences from the conventional method of allocating 3G or 4G spectrum to operators. First, because the availability of LSA spectrum will be determined by the usage status of the spectrum owner, the configuration needs to work in a shorter time (perhaps several times per hour), the regulatory bodies or public platforms allocate the spectrum under the LSA licensing mechanism by auctions, and once the spectrum owner releases his free spectrum, the spectrum will be streamed to the public spectrum pool for secondary streaming, which can greatly improve the efficiency of spectrum usage. Secondly, by utilizing the space reuse characteristic of the frequency spectrum (secondary frequency spectrum buyers which cannot cause interference can use the same frequency spectrum band at the same time), according to the geographic and use information reported by the secondary frequency spectrum buyers, an auction platform draws a frequency-use interference relation diagram between the governed coverage areas, and a plurality of operating base stations which are not interfered with each other and have similar frequency spectrum requirements are distributed to form a group purchasing group. If the group purchase is successful, the spectrum allocated to a group may be used by all members of the group.

Claims

1. The bilateral dynamic multi-demand ascending auction spectrum allocation method based on the cloud group purchase platform is characterized by comprising the following steps: comprises the following steps of (a) carrying out,

s1, the frequency spectrum auction system comprises three parties: sellers, buyers and cloud buying a spectrum platform; the cloud group purchase spectrum platform receives all spectrum resources reported by a seller and corresponding spectrum resource quotations, divides all spectrum resources into k spectrum resource blocks, and receives respective bids B of the buyer on the k spectrum resource blocks_iBid price B_iThe bidding values are arranged from high to low, the cloud group purchase frequency spectrum platform divides all buyers into h group purchase groups, and obtains group purchase group bidding vectors of each group purchase group for k frequency spectrum resource blocks

s34, repeating the steps S31-S33, and calculating the actual payment values of the spectrum resource blocks distributed by all buyers in all group purchase groups;

and S4, according to the actual payment value of each buyer of the same spectrum resource block obtained in the step S3, comparing the total price of all spectrum resource blocks belonging to the same spectrum resource calculated by the cloud-based spectrum-buying platform with the frequency spectrum resource quotation of the seller of the spectrum resource, paying for bargaining when the total price is higher than the frequency spectrum resource quotation, and streaming when the total price is lower than the frequency spectrum resource quotation.

2. The cloud buying platform based two-sided dynamic multi-demand ascending auction spectrum allocation method of claim 1, wherein: the group buying group is divided according to the geographical location information and the maximum transmitting power of all the buyers.

3. The cloud buying platform based two-sided dynamic multi-demand ascending auction spectrum allocation method of claim 1, wherein: the specific implementation method of step S2 is:

Counting the bids of all the spectrum resource blocks in the group buying group, obtaining the number of resources which can be shot in each group buying group in turn, and removing one group buying group G in turn_hCalculating the total number of the clapable resources of the rest group purchase groups and the total number of the clapable resources of all the group purchase groups;

s2.4, if the total number of the resources which can be shot of one remaining group purchase group is less than the k value and the total number of the resources which can be shot of all the group purchase groups is more than the k value, the cloud group purchase spectrum platform divides one spectrum resource block to provide the G removed in the cloud group purchase spectrum platform_h' taking the p value of the current round as the p value of the next round, subtracting 1 from the k value, continuing to call the price, and repeating the step S2.2;