CN104378740A - Trunking communication calling system and method - Google Patents

Abstract

The invention discloses a trunking communication calling method and system, in particular to a trunking calling method and system applied to intellectual property networks. Communication between users in different groups can be rapidly achieved across regions and across majors, and multiple functions such as announcement making, group meetings, across-group discussion and consultation responding can be achieved; meanwhile, the dynamic allocation arithmetic in the prior art can be improved, and the using rate and the response balance of channels can be further improved.

Description

A kind of colony communication calling system and method

Technical field

The present invention relates to communication technical field, be specifically related to a kind of group calling system and the method that are applied to intellectual property network.

Background technology

Trunked communication system, be a kind of specialized mobile radio technology, it shares the dynamic assignment with channel by multichannel, thus effectively utilizes frequency resource and Internet resources, thus reduces construction and the operation cost of communication network.

First generation analog cluster system, the resource of shared frequencies, equipment, service, can realize centralized maintenance and the management of system.Adopt simplex mode, utilize multichannel technology of sharing to achieve " message trunking " and " transmission trunking ", improve channel utilance.Second generation digital cluster system, the availability of frequency spectrum greatly improves than analogue system, has expanded power system capacity further.Trunked communication system adopts the half-duplex call of one-to-many usually, first must obtain right of speech before group member speech, and only have acquisition right of speech just to allow speech, the group member not obtaining right of speech can only be selected to answer.Group service is the communication service between group user, this just needs to manage group: must set up in advance and select a group be made up of one or more group members when calling is initiated, then the group calling of this group is just initiated by PTT (Push-to-Talk, PTT) the group calling key in terminal.In third generation partner program (3GPP) agreement, when main dialing user initiates group calling in relay MSC, relay MSC obtains the address of controlling MSC from Group Call Register (GCR), and select interoffice route accordingly, use group call reference to construct call setup message as called number, be sent to controlling MSC.

At present, wireless cluster communication system has been widely used in the command dispatching and communication system of all trades and professions and various occasion.For intellectual property industry, owing to having stronger regionality and specialty division, also there is not the comprehensive group calling system of nationwide integrated patent, copyright and trade mark at present.

Summary of the invention

The present invention, in conjunction with the feature in intellectual property field, provides a kind of colony communication calling system and method, is particularly applied to group calling system and the method for intellectual property network.This group calling system and method can cross over area and field, communication is realized fast in the user of different grouping, and can realize making an announcement, colony's meeting, discuss across group, the several functions such as consulting response, the algorithm of the present invention simultaneously to dynamic assignment of the prior art is improved, and the utilance and the response that improve channel are further balanced.

The object of the invention is to be achieved through the following technical solutions:

A kind of group calling system, comprise set of packets groupcall unit, this set of packets groupcall unit comprises audio call module, sends call instruction, the interactive voice that multiple calling party is simultaneously online can be realized, and comprise voice broadcast service module, send broadcasting instructions, unidirectional transmission voice messaging can be realized, this group calling system also comprises allocation units, receive the information that set of packets groupcall unit sends, if be call instruction, then call information is sent to buffer; If be broadcasting instructions, then directly to caller party terminal pushed information.

A kind of intellectual property network cluster calling system, comprise calling party's grouped element, described calling party's grouped element can multi-step grouping, one-level is grouped into divides into groups according to patent, copyright, trade mark, wherein, patent secondary data packets is set up in the grouping of patent one-level further, comprises invention, utility model, outward appearance grouping, and can be set by the user according to region or type of service Further Division, each grouping all has unique ID.

Preferably, described calling party's grouped element has adjustable, according to region zones, can comprise South China, East China, North China and national each province, each municipality directly under the Central Government further, and can cross over level according to setting and set up new group; Divide according to type of service, multiple traffic packets such as copyright consulting, dispute over copyright, trade mark registration, trademark assignment can be comprised further, and above-mentioned grouping also can according to setting, in different groupings, selected associated user sets up new grouping, to solve the situation of cross-cutting interchange when there is cross business, while above-mentioned grouping dynamically updates, change packet ID.

Preferably, in group mode, single variable value is had to divide into groups and group distance grouping two kinds, wherein, single variable value grouping is using a variate-value as one group, and group is that whole variate-value is divided into some intervals successively apart from grouping, and using the variate-value in each interval as one group.

Preferably, also comprise set of packets groupcall unit, comprise audio call module, send call instruction, the interactive voice that multiple calling party is simultaneously online can be realized, and comprise voice broadcast service module, send broadcasting instructions, unidirectional transmission voice messaging can be realized; And

Allocation units, receive the information that set of packets groupcall unit sends, if be call instruction, then call information are sent to buffer; If be broadcasting instructions, then directly to caller party terminal pushed information;

Buffer memory unit, extracts the packet ID entrained by different call request information, by the same packets ID duplicate removal of the different calling parties in special time period, and to different grouping ID with time order and function and priority orders, and arrangement buffer memory;

Memory cell, stored calls user registers parameter information, comprises terminal name, IP address of terminal, packet ID, and obtains renewal in real time according to the update request of calling party place group, according to dynamic assignment principle, between corresponding packet ID and multiple divisional processing device, set up mapping relations;

Control unit, packet ID in reception buffer successively, and according to the call request information that the mapping relations that call register unit stores specify divisional processing device to receive corresponding calling party to send, ensure the calling of same group member, use the divisional processing device process of specifying;

Multiple divisional processing device unit, sends instruction according to the call request information that calling party sends, is sent to called party's terminal, and according to called party's terminal feedback information, sets up corresponding communication channel.

Preferably, also comprise dynamic assignment divisional processing device unit, it has solicited message process balance module, makes the work relative equilibrium of each divisional processing device.

Preferably, dynamic assignment processor unit comprises dynamic channel allocation unit, it comprises following modules: (1) channel distribution module: all groupings of different I D are divided into channel N1, N2 ... add up to N, then grouped channels excursion is 0 ~ N, and concrete distribution condition, by the traffic and conflict restriction, represents conflict restriction with matrix X:

$X=\left|\begin{array}{ccc}{X}_{11}& ...& X_{1N}\\ ...& X_{\mathrm{ij}}& ...\\ X_{M1}& ...& X_{\mathrm{MN}}\end{array}\right|$

Wherein, X _ijrepresent the conflict restriction of channel, i and j represents the row and column in place conflict restriction matrix respectively, and M is whole channel number, when i and j that divide into groups can not reuse a channel, and X _ij=1, if when grouping i and j can reuse a channel, X _ij=0, each grouped channels service condition matrix A represents: $A=\left|\begin{array}{ccc}{a}_{11}& ...& a_{1N}\\ ...& a_{\mathrm{ik}}& ...\\ a_{M1}& ...& a_{\mathrm{MN}}\end{array}\right|$

Wherein, if grouping i allocated channel, then a is had _ik=1, otherwise, a _ik=0; The method comprises the steps:

Every a line by matrix A is mapped as independently goes here and there, and namely the i-th row is mapped as A _i=a _i1a _i2... a _iN, whole system channel B is mapped as M*N position and independently goes here and there, and version is B=a ₁₁a ₁₂... a _1N... a _i1a _i2... a _iN... a _m1a _m2... a _mN, the wherein corresponding gene strand of each state;

(2) initial population generation module: setting initial population individual amount is 30, and stochastic generation some is individual, is the multiple of 30, therefrom screens defect individual and puts in initialization population, until population at individual quantity reaches requirement;

(3) the fitness function constructing module of grouped channels: setting t _ibe the number of channel of i-th grouping demand, then have fitness function:

$f\left(B\right)=\underset{i=1}{\overset{M}{\mathrm{\Σ}}}{(M-|t_i-\underset{K=1}{\overset{N}{\mathrm{\Σ}}}{a}_{\mathrm{ik}}\left|\right)}^2+\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{K=1}{\overset{N}{\mathrm{\Σ}}}\underset{i\≠j}{\overset{M}{\underset{j=1}{\mathrm{\Σ}}}}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}(1-X_{\mathrm{ij}})a_{\mathrm{iK}}{a}_{\mathrm{jK}}$

Wherein, aiK is the i-th row K row channel usage, and ajK is jth row K row channel usages;

(4) to the selection operational module of different I D grouping: selection operation is on the basis of ideal adaptation degree assessment in all groupings being based upon multiple different I D, in assumed group, certain individual B ' fitness is f (B '), then certain individual B ' is P by the probability selected _s=f (B ')/Σ f (B '), ideal adaptation degree is larger, higher by the probability selected, and ideal adaptation degree is less, lower by select probability, so just can determine that those individualities are selected;

(5) channel assignment module, it is according to genetic operator, determines final Channel distribution mode, and wherein, genetic operator comprises the crossover operator of improvement and the mutation operator of improvement, the crossover probability P of the crossover operator wherein improved _cchange formula is:

$P_c=\left\{\begin{array}{cc}{P}_{c2}-\frac{(P_{c2}-P_{c1})\×(f_{\mathrm{avg}}-f_i)}{(f_{\mathrm{avg}}-f_{\min })}& f_i\≤f_{\mathrm{avg}}\\ P_{c3}-\frac{(P_{c3}-P_{c2})\×(f_{\max }-f_i)}{(f_{\max }-f_{\mathrm{avg}})}& f_i>f_{\mathrm{avg}}\end{array}\right.$

Wherein, f _avgfor the whole individual average fitness value of contemporary population, f _maxfor maximum adaptation angle value individual in population, f _minfor minimum fitness value individual in population, f _ibe that in 2 intersection individualities, fitness value is larger, P _c1, P _c2, P _c3represent the value of three different phase crossover probabilities respectively, P _c1represent f _icrossover probability when=0, P _c2represent f _i=f _avgtime crossover probability, P _c3represent f _i=f _maxtime crossover probability;

The mutation probability P of the mutation operator improved _mformula as follows:

$P_m=\left\{\begin{array}{cc}{P}_{m1}& f_i\≤f_{\mathrm{avg}}\\ P_{m2}-\frac{(P_{m2}-P_{m1})\×(f_{\max }-f_i)}{(f_{\max }-f_{\mathrm{avg}})}& f_i>f_{\mathrm{avg}}\end{array}\right.$

Wherein f _ifor chromosome fitness value, f _avgfor population average fitness value, P _m1, P _m2represent the numerical value of two different phase mutation probability respectively, P _m1represent f _i≤ f _avgtime mutation probability, P _m2represent f _i> f _avgtime mutation probability;

End condition, when algorithm evolution is to regulation algebraically, such as, when being 100 generation, the solution that the individual B ' that fitness value f (B ') is corresponding finally finds for algorithm.

Adopt technical scheme of the invention process, voice, broadcast capability can be focused on, and ensure the calling of same group member, by the process of appointment divisional processing device, thus save system resource, and good experience can be brought to user, reduce repeated call number of times.

Accompanying drawing explanation

Fig. 1 is the intellectual property network cluster calling system packet diagram according to the embodiment of the present invention;

Fig. 2 is the intellectual property network cluster calling system composition frame chart according to the embodiment of the present invention;

Fig. 3 is the dynamics of channels allocation flow figure according to the embodiment of the present invention;

Fig. 4 is the intellectual property network cluster calling system call flow diagram according to the embodiment of the present invention.

Embodiment

Understand for the ease of persons skilled in the art and realize the present invention, now describing example of the present invention by reference to the accompanying drawings.

Intellectual property calling terminal, has multiple block form, wherein, as a part of example, can as shown in Figure 1, and carry out lower column split, one-level is grouped into: patent group, copyright group and trade mark group; Secondary data packets is: patent group is further divided into outward appearance group, utility model group and invention group.Arrange according to user simultaneously, according to geographic location area and type of service, again divide into groups, such as can be divided into according to region: Beijing group, Shanghai group, Hebei group etc., can combine before any group; Can be divided into according to traffic packets: patent information issues bulletin group, patent consultation group, application process inquiry group, patent right transfer interchange group etc.

The grouping of intellectual property calling terminal, has adjustable, according to region zones, can comprise South China, East China, North China and national each province, each municipality directly under the Central Government further, and can cross over level according to artificial setting and set up new group; Divide according to type of service, multiple traffic packets such as copyright consulting, dispute over copyright, trade mark registration, trademark assignment can be comprised further, and above-mentioned grouping also can according to artificial setting, in different groupings, selected associated user sets up new grouping, to solve the situation of cross-cutting interchange when there is cross business.The equal dynamic of above-mentioned grouping upgrades, and changes packet ID simultaneously.

In group mode, single variable value is had to divide into groups and group distance grouping two kinds.Single variable value grouping is using a variate-value as one group, and this grouping general value is applicable to discrete variable, and uses in the situation that variate-value is less.When continuous variable or variate-value more, usual employing group is apart from grouping.It is that whole variate-value is divided into some intervals successively, and using the variate-value in this interval as one group.In group apart from grouping, the minimum value of a group is called lower limit; The maximum of a group is called the upper limit.When employing group is apart from grouping, need not follow heavyly the principle of not leaking.Heavily do not refer to that an item number is according to a certain group that can only divide wherein, can not repeat in other groups; Do not leak and refer to that group can limit, namely in divided whole groups, each item number, according to a certain group that can divide wherein, can not be omitted.Group number generally with the feature of data itself and data number relevant.One of object due to grouping is the feature of observed data distribution, therefore organize number number should be moderate.If group number very little, the distribution of data will be too concentrated, and group number is too many, and Data distribution8 will too disperse, and this is all not easy to the characteristic sum rule of observed data distribution.The determination of group number should for the purpose of the distribution characteristics that can show data and rule.During practical application, can according to data number and feature and analyze requirement determine organize number.Determine the group distance of each group.Group is apart from being the upper limit of a group and the difference of lower limit.Group apart from can according to the maximum of total data and minimum value and the group number that divide determine, and group apart from=(maximum one minimum value)/organize number.Frequency distribution table is organized into according to grouping.

By packet transaction, the processing speed of mass data can be improved, and by the standardized designs of packet, thus achieve the dynamic equalization of packet.

Embodiment one

Present embodiment describes a kind of intellectual property network cluster calling system, as shown in Figure 2, this system comprises:

Calling party's grouped element 100, have patent group 101, copyright group 102, trade mark group 103, wherein, patent group also comprises invention group 1011, utility model group 1012, outward appearance group 1013, and can be set by the user according to region or business Further Division, each grouping all has unique ID;

Set of packets groupcall unit 200, comprises audio call module 201, sends call instruction, can realize the interactive voice that multiple calling party is simultaneously online, and comprise voice broadcast service module 202, send broadcasting instructions, can realize unidirectional transmission voice messaging;

Allocation units 300, receive the information that set of packets groupcall unit sends, if be call instruction, then call information are sent to buffer 400; If be broadcasting instructions, then directly to caller party terminal pushed information;

Buffer memory unit 400, extracts the packet ID entrained by different call request information, by the same packets ID duplicate removal of the different calling parties in special time period, and to different grouping ID with time order and function and priority orders, and arrangement buffer memory;

Memory cell 500, stored calls user registers parameter information, comprise terminal name, IP address of terminal, packet ID, and obtain renewal in real time according to the update request of calling party place group, according to dynamic assignment principle, between corresponding packet ID and multiple divisional processing device 700, set up mapping relations;

Control unit 600, packet ID in reception buffer successively, and according to the call request information that the mapping relations that call register unit stores specify divisional processing device to receive corresponding calling party to send, ensure the calling of same group member, use the divisional processing device process of specifying;

Multiple divisional processing device unit 700, sends instruction according to the call request information that calling party sends, is sent to called party's terminal, and according to called party's terminal feedback information, sets up corresponding communication channel;

Dynamic assignment divisional processing device unit 800, according to solicited message process homeostatic principle, makes the work relative equilibrium of each divisional processing device.

The calling of mobile communication system is all point-to-point communication mode, to connect and Traffic Channel can meet call caller and called call demand as long as system sets up a pair signaling, system for Traffic Channel that user provides be all converse in user special.And the calling of trunked communication system, be all point-to-multipoint mostly, Traffic Channel is all shared, and at this moment calling subscriber is one, and called subscriber has a lot of, and everybody monitors a Traffic Channel, and called subscriber is not probably distributed in a region.This makes system will set up signaling for each region to connect and shared service channel, simultaneously also will according to the movement of group user, real-time change the region of service is provided, group user to new region set up signaling connection and shared service channel.The mode of this shared channel of trunked communication system, than the mode of the dedicated channel of mobile communication, more complex, need to rethink that this algorithm of channel resource allocation algorithm is process more consuming time.The application also will solve the Channel Assignment Problems of trunked communication system.

Dynamic assignment processor unit comprises dynamic channel allocation unit, it comprise channel distribution module, initial population generation module, grouped channels fitness function constructing module, to different I D grouping selection operational module and channel assignment module, its workflow is as shown in Figure 3: channel distribution module, by all groupings of different I D processed in group system, be divided into channel N1, N2 ... add up to N, then grouped channels excursion is 0 ~ N, concrete distribution condition, by the traffic and conflict restriction, represents conflict restriction with matrix X:

$X=\left|\begin{array}{ccc}{X}_{11}& ...& X_{1N}\\ ...& X_{\mathrm{ij}}& ...\\ X_{M1}& ...& X_{\mathrm{MN}}\end{array}\right|$

Wherein, X _ijrepresent the conflict restriction of channel, i and j represents the row and column in place conflict restriction matrix respectively, and M is whole channel number, when i and j that divide into groups can not reuse a channel, and X _ij=1, if when grouping i and j can reuse a channel, X _ij=0, each grouped channels service condition matrix A represents: $A=\left|\begin{array}{ccc}{a}_{11}& ...& a_{1N}\\ ...& a_{\mathrm{ik}}& ...\\ a_{M1}& ...& a_{\mathrm{MN}}\end{array}\right|$

Wherein, if grouping i allocated channel, then a is had _ik=1, otherwise, a _ik=0; The method comprises the steps:

Every a line by matrix A is mapped as independently goes here and there, and namely the i-th row is mapped as A _i=a _i1a _i2... a _iN, whole system channel B is mapped as M*N position and independently goes here and there, and version is B=a ₁₁a ₁₂... a _1N... a _i1a _i2... a _iN... a _m1a _m2... a _mN, the wherein corresponding gene strand of each state;

Be gene by the Parameter Switch in Channel Assignment Problems, be specially and independently gone here and there a grouped channels service condition N position, the every a line by matrix A is mapped as independently goes here and there, i.e. A _i=a _i1a _i2... a _iN, whole system channel mapping is that M*N position is independently gone here and there, and version is B=a ₁₁a ₁₂... a _1N... a _i1a _i2... a _iN... a _m1a _m2... a _mN, the wherein corresponding gene strand of each state.

Initial population generation module: setting initial population individual amount is 30, stochastic generation some individualities (being the multiple of 30), therefrom screens defect individual and puts in initialization population, until population at individual quantity reaches requirement.

The fitness function structure of grouped channels, genetic algorithm take fitness function as foundation, and channel allocation, by the traffic and conflict restriction, sets t _ibe the number of channel of i-th grouping demand, then have fitness function:

$f\left(B\right)=\underset{i=1}{\overset{M}{\mathrm{\Σ}}}{(M-|t_i-\underset{K=1}{\overset{N}{\mathrm{\Σ}}}{a}_{\mathrm{ik}}\left|\right)}^2+\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{K=1}{\overset{N}{\mathrm{\Σ}}}\underset{i\≠j}{\overset{M}{\underset{j=1}{\mathrm{\Σ}}}}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}(1-X_{\mathrm{ij}})a_{\mathrm{iK}}{a}_{\mathrm{jK}}$

Wherein, aiK is the i-th row K row channel usage, and ajK is jth row K row channel usages;

Selection operation to different I D grouping: selection operation is based upon on the basis of individual in population Fitness analysis, and in assumed group, certain individual B ' fitness is f (B '), then certain individual B ' is P by the probability selected _s=f (B ')/Σ f (B '), ideal adaptation degree is larger, higher by the probability selected, and ideal adaptation degree is less, lower by select probability.So just can determine that those individualities are selected.

According to genetic operator, determine final Channel distribution mode:

Genetic operator comprises the crossover operator of improvement and the mutation operator of improvement, the crossover probability P of the crossover operator wherein improved _cchange formula is:

$P_c=\left\{\begin{array}{cc}{P}_{c2}-\frac{(P_{c2}-P_{c1})\×(f_{\mathrm{avg}}-f_i)}{(f_{\mathrm{avg}}-f_{\min })}& f_i\≤f_{\mathrm{avg}}\\ P_{c3}-\frac{(P_{c3}-P_{c2})\×(f_{\max }-f_i)}{(f_{\max }-f_{\mathrm{avg}})}& f_i>f_{\mathrm{avg}}\end{array}\right.$

Wherein, f _avgfor the whole individual average fitness value of contemporary population, f _maxfor maximum adaptation angle value individual in population, f _minfor minimum fitness value individual in population, f _ibe that in 2 intersection individualities, fitness value is larger, P _c1, P _c2, P _c3represent the value of three different phase crossover probabilities respectively, P _c1represent f _icrossover probability when=0, P _c2represent f _i=f _avgtime crossover probability, P _c3represent f _i=f _maxtime crossover probability;

The mutation probability P of the mutation operator improved _mformula as follows:

$P_m=\left\{\begin{array}{cc}{P}_{m1}& f_i\≤f_{\mathrm{avg}}\\ P_{m2}-\frac{(P_{m2}-P_{m1})\×(f_{\max }-f_i)}{(f_{\max }-f_{\mathrm{avg}})}& f_i>f_{\mathrm{avg}}\end{array}\right.$

Wherein f _ifor chromosome fitness value, f _avgfor population average fitness value.P _m1, P _m2represent the numerical value of two different phase mutation probability respectively, P _m1represent f _i≤ f _avgtime mutation probability, P _m2represent f _i> f _avgtime mutation probability.

End condition, when algorithm evolution is to regulation algebraically, such as, during 100 generation, the individual B that fitness value f (B) is corresponding is the solution that algorithm finally finds.

This embodiment improves the performance of trunked call, make KPI (Key Performance Indications), namely refer to that KPI Key Performance Indicator obtains larger lifting, be in particular in following index: (1) call completing rate, call completing rate >=95% of the present embodiment, compared with the call completing rate 80% of industry standard, higher than industry standard 15%, thus improve customer satisfaction.(2) service level, namely in certain timing statistics section, in some seconds, response telephone number amount and call center access the percentage of phone, adopt the present embodiment, the phone of 95% made response within 20 seconds, and the phone that industry standard is 80% made response within 20 seconds.(3) average queuing time, namely in certain section of timing statistics, the average waiting duration of the wait of awaiting acknowledgement after caller shortlists.Average queuing time≤10 second of the present embodiment, and industry standard is 20 seconds, substantially reduces queue time.Therefore, invention increases call completing rate, reduce the stand-by period, and queuing time, cost is significantly declined.

Embodiment two

Trunked call is all with the common busy channel of group cluster user in coverage.Usually each user will take the independently communication resource, also just means that resource occupation has how many users to enter group calling when business realizing, will take how many channel resources, also will double for resource occupation PTT.

For this prior art problem, present embodiment describes a kind of workflow of intellectual property network cluster calling system, by optimizing the workflow of group calling system, thus saved the communication resource, and achieve and set up communication channel fast, its specific works flow process as shown in Figure 4, step 1, calling party's grouped element sends audio call or broadcasting instructions, audio call can realize the simultaneously online interactive voice of multiple calling party, broadcasting instructions, can realize unidirectional transmission voice messaging; Step 2, the information that allocation units receipt of call user grouping unit sends, and the type of information is identified, if recognition result is call instruction, information is sent to buffer memory unit buffer memory, if recognition result is broadcasting instructions, then information is directly pushed to user terminal; Step 3, buffer memory unit processes the information received, and extracts the packet ID that solicited message is carried, in special time period, by the same packets ID duplicate removal of different calling party, and to different grouping ID with time order and function and preset priority orders sequence; Step 4, utilizes the customer parameter information stored in memory cell, and the mapping relations table between packet ID and processor unit, and call request information is distributed to different divisional processing device unit by control unit; Step 5, according to called party's terminal feedback information, is confirmed whether to allow to connect, if allowed, sets up communication channel.

Utilize the channel allocation of the trunked communication system of the same embodiment one: dynamic assignment processor unit comprises dynamic channel allocation unit simultaneously, it comprises channel distribution module, initial population generation module, the fitness function constructing module of grouped channels, to selection operational module and the channel assignment module of different I D grouping, its workflow is as shown in Figure 3: channel distribution module, by all groupings of different I D processed in group system, be divided into channel N1, N2 add up to N, then grouped channels excursion is 0 ~ N, concrete distribution condition is by the traffic and conflict restriction, conflict restriction is represented with matrix X:

$X=\left|\begin{array}{ccc}{X}_{11}& ...& X_{1N}\\ ...& X_{\mathrm{ij}}& ...\\ X_{M1}& ...& X_{\mathrm{MN}}\end{array}\right|$

Wherein, X _ijrepresent the conflict restriction of channel, i and j represents the row and column in place conflict restriction matrix respectively, and M is whole channel number, when i and j that divide into groups can not reuse a channel, and X _ij=1, if when grouping i and j can reuse a channel, X _ij=0, each grouped channels service condition matrix A represents: $A=\left|\begin{array}{ccc}{a}_{11}& ...& a_{1N}\\ ...& a_{\mathrm{ik}}& ...\\ a_{M1}& ...& a_{\mathrm{MN}}\end{array}\right|$

Wherein, if grouping i allocated channel, then a is had _ik=1, otherwise, a _ik=0; The method comprises the steps:

Every a line by matrix A is mapped as independently goes here and there, and namely the i-th row is mapped as A _i=a _i1a _i2... a _iN, whole system channel B is mapped as M*N position and independently goes here and there, and version is B=a ₁₁a ₁₂... a _1N... a _i1a _i2... a _iN... a _m1a _m2... a _mN, the wherein corresponding gene strand of each state;

Be gene by the Parameter Switch in Channel Assignment Problems, be specially and independently gone here and there a grouped channels service condition N position, the every a line by matrix A is mapped as independently goes here and there, i.e. A _i=a _i1a _i2... a _iN, whole system channel mapping is that M*N position is independently gone here and there, and version is B=a ₁₁a ₁₂... a _1N... a _i1a _i2... a _iN... a _m1a _m2... a _mN, the wherein corresponding gene strand of each state.

Initial population generation module: setting initial population individual amount is 30, stochastic generation some individualities (being the multiple of 30), therefrom screens defect individual and puts in initialization population, until population at individual quantity reaches requirement.

The fitness function structure of grouped channels, genetic algorithm take fitness function as foundation, and channel allocation, by the traffic and conflict restriction, sets t _ibe the number of channel of i-th grouping demand, then have fitness function:

$f\left(B\right)=\underset{i=1}{\overset{M}{\mathrm{\Σ}}}{(M-|t_i-\underset{K=1}{\overset{N}{\mathrm{\Σ}}}{a}_{\mathrm{ik}}\left|\right)}^2+\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{K=1}{\overset{N}{\mathrm{\Σ}}}\underset{i\≠j}{\overset{M}{\underset{j=1}{\mathrm{\Σ}}}}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}(1-X_{\mathrm{ij}})a_{\mathrm{iK}}{a}_{\mathrm{jK}}$

Wherein, aiK is the i-th row K row channel usage, and ajK is jth row K row channel usages;

Selection operation to different I D grouping: selection operation is based upon on the basis of individual in population Fitness analysis, and in assumed group, certain individual B ' fitness is f (B '), then certain individual B ' is P by the probability selected _s=f (B ')/Σ f (B '), ideal adaptation degree is larger, higher by the probability selected, and ideal adaptation degree is less, lower by select probability.So just can determine that those individualities are selected.

According to genetic operator, determine final Channel distribution mode:

Genetic operator comprises the crossover operator of improvement and the mutation operator of improvement, the crossover probability P of the crossover operator wherein improved _cchange formula is:

$P_c=\left\{\begin{array}{cc}{P}_{c2}-\frac{(P_{c2}-P_{c1})\×(f_{\mathrm{avg}}-f_i)}{(f_{\mathrm{avg}}-f_{\min })}& f_i\≤f_{\mathrm{avg}}\\ P_{c3}-\frac{(P_{c3}-P_{c2})\×(f_{\max }-f_i)}{(f_{\max }-f_{\mathrm{avg}})}& f_i>f_{\mathrm{avg}}\end{array}\right.$

Wherein, f _avgfor the whole individual average fitness value of contemporary population, f _maxfor maximum adaptation angle value individual in population, f _minfor minimum fitness value individual in population, f _ibe that in 2 intersection individualities, fitness value is larger, P _c1, P _c2, P _c3represent the value of three different phase crossover probabilities respectively, P _c1represent f _icrossover probability when=0, P _c2represent f _i=f _avgtime crossover probability, P _c3represent f _i=f _maxtime crossover probability;

The mutation probability P of the mutation operator improved _mformula as follows:

$P_m=\left\{\begin{array}{cc}{P}_{m1}& f_i\≤f_{\mathrm{avg}}\\ P_{m2}-\frac{(P_{m2}-P_{m1})\×(f_{\max }-f_i)}{(f_{\max }-f_{\mathrm{avg}})}& f_i>f_{\mathrm{avg}}\end{array}\right.$

Wherein f _ifor chromosome fitness value, f _avgfor population average fitness value.P _m1, P _m2represent the numerical value of two different phase mutation probability respectively, P _m1represent f _i≤ f _avgtime mutation probability, P _m2represent f _i> f _avgtime mutation probability.

End condition, when algorithm evolution is to regulation algebraically, such as, during 100 generation, the individual B that fitness value f (B) is corresponding is the solution that algorithm finally finds.

This embodiment optimizes the workflow of trunked call, make KPI (Key Performance Indications), namely refer to that KPI Key Performance Indicator obtains larger lifting, be in particular in following index: (1) shortens turn-on time, the present embodiment turn-on time≤15 seconds, compared with industry standard≤30 second, period of reservation of number shortens half, thus improves operating efficiency; (2) avoiding repeating dialing, by adopting the workflow of the present embodiment, making same point of group member, avoid during intercommunication mutually repeating dialing, focused on by same divisional processing device, thus connect fast; (3) communication resources utilization ratio, the present embodiment is by the mapping between packet ID and divisional processing device, take full advantage of the communication resource, communication resources utilization ratio is made to reach 99%, higher than the requirement of current industry standard 85%, thus improve the utilance of the communication resource, also reduce the maintenance cost of group calling system; (4) by the mode that dynamic assignment is shared, decrease the time delay in network processes calling procedure, have access fast, high channel efficiency and frequency spectrum utilization rate, higher user privacy, expansibility and the technological merit such as supporting business kind is many.Channel sharing can regulate supplies, and the number of users summation that total number of users that dynamic assignment channel can be supported can be supported than each self-channel that these total channels during dispersion networking are distributed to each net is much bigger, thus also can improve service quality; Concentrated networking can also strengthen management and safeguard, thus can improve the grade of service, enhance system function.

Although depict the present invention by embodiment, those of ordinary skill in the art know, without departing from the spirit and substance in the present invention, the present invention just can be made to have many distortion and change, and scope of the present invention is limited to the appended claims.

Claims (7)

1. a group calling system, comprise set of packets groupcall unit, this set of packets groupcall unit comprises audio call module, it sends call instruction, realize the interactive voice that multiple calling party is simultaneously online, this set of packets groupcall unit also comprises voice broadcast service module, sends broadcasting instructions, realize unidirectional transmission voice messaging, it is characterized in that:

Also comprise allocation units, receive the information that set of packets groupcall unit sends, described information comprises call instruction and broadcasting instructions, if be call instruction, then call information is sent to buffer buffer memory; If be broadcasting instructions, then directly to caller party terminal pushed information.

2. an intellectual property network cluster calling system, comprise calling party's grouped element, it is characterized in that: described calling party's grouped element, it has multi-step grouping, one-level grouping comprises patent, copyright, trade mark grouping, under the grouping of patent one-level, set up patent secondary data packets, comprises invention, utility model, outward appearance grouping, and can be set by the user and again divide into groups according to region or type of service, every grade of each grouping all has unique ID.

3. an intellectual property network cluster calling system as claimed in claim 2, it is characterized in that: described calling party's grouped element is according to region zones, South China, East China, North China can be comprised, and set up new grouping according to setting leap level; Divide according to type of service, comprise copyright consulting, dispute over copyright, trade mark registration, the multiple traffic packets of trademark assignment, and above-mentioned grouping can be selected associated user and sets up new grouping in different groupings, during above-mentioned grouping renewal, change packet ID simultaneously.

4. an intellectual property network cluster calling system as claimed in claim 3, it is characterized in that: grouping comprises single variable value grouping and the grouping of group distance, single variable value grouping is as one group using a variate-value, group is that whole variate-value is divided into some intervals successively apart from grouping, and using the variate-value in each interval as one group.

5. an intellectual property network cluster calling system as claimed in claim 4, it is characterized in that: also comprise set of packets groupcall unit, comprise audio call module, send call instruction, the interactive voice that multiple calling party is simultaneously online can be realized, and comprise voice broadcast service module, send broadcasting instructions, unidirectional transmission voice messaging can be realized; It is characterized in that: also comprise

Allocation units, receive the information that set of packets groupcall unit sends, described information comprises call instruction and broadcasting instructions, if be call instruction, then call information is sent to buffer buffer memory; If be broadcasting instructions, then directly to caller party terminal pushed information; Buffer memory unit, extracts the packet ID entrained by different call request information, by the same packets ID duplicate removal of the different calling parties in special time period, and to different grouping ID with time order and function and the priority orders that presets, arranges the ID of buffer memory;

Memory cell, stored calls user registers parameter information, comprises terminal name, IP address of terminal, packet ID, and obtains renewal in real time according to the update request of calling party place group, according to dynamic assignment principle, between corresponding packet ID and multiple divisional processing device, set up mapping relations;

Control unit, packet ID in reception buffer successively, and according to the call request information that the mapping relations that call register unit stores specify divisional processing device to receive corresponding calling party to send, ensure the calling of same group member, use the divisional processing device process of specifying;

Multiple divisional processing device unit, sends instruction according to the call request information that calling party sends, is sent to called party's terminal, and according to called party's terminal feedback information, sets up corresponding communication channel.

6. an intellectual property network cluster calling system as claimed in claim 5, characterized by further comprising:

Dynamic assignment divisional processing device unit, it has solicited message process balance module, makes the work relative equilibrium of each divisional processing device.

7. an intellectual property network cluster calling system as claimed in claim 6, is characterized in that: dynamic assignment processor unit comprises dynamic channel allocation unit, and it comprises following modules:

(1) channel distribution module: all groupings of different I D are divided into channel N1, N2 ..., add up to N, then grouped channels excursion is 0 ~ N, and concrete distribution condition, by the traffic and conflict restriction, represents conflict restriction with matrix X:

$X=\left|\begin{array}{ccc}{M}_{11}& ...& M_{1N}\\ ...& M_{\mathrm{ij}}& ...\\ M_{M1}& ...& M_{\mathrm{MN}}\end{array}\right|$

Wherein, X _ijrepresent the conflict restriction of channel, i and j represents the row and column in place conflict restriction matrix respectively, and M is whole channel number, when i and j that divide into groups can not reuse a channel, and X _ij=1, if when grouping i and j can reuse a channel, X _ij=0, each grouped channels service condition matrix A represents: $A=\left|\begin{array}{ccc}{a}_{11}& ...& a_{1N}\\ ...& a_{\mathrm{ik}}& ...\\ a_{M1}& ...& a_{\mathrm{MN}}\end{array}\right|$

Wherein, if grouping i allocated channel, then a is had _ik=1, otherwise, a _ik=0; The method comprises the steps:

Every a line by matrix A is mapped as independently goes here and there, and namely the i-th row is mapped as A _i=a _i1a _i2... a _iN, whole system channel B is mapped as M*N position and independently goes here and there, and version is B=a ₁₁a ₁₂... a _1N... a _i1a _i2... a _iN... a _m1a _m2... a _mN, the wherein corresponding gene strand of each state;

(2) initial population generation module: setting initial population individual amount is 30, and stochastic generation some is individual, is the multiple of 30, therefrom screens defect individual and puts in initialization population, until population at individual quantity reaches requirement;

(3) the fitness function constructing module of grouped channels: setting t _ibe the number of channel of i-th grouping demand, then have fitness function:

$f\left(B\right)=\underset{i=1}{\overset{M}{\mathrm{\Σ}}}{(M-|t_i-\underset{K=1}{\overset{N}{\mathrm{\Σ}}}{a}_{\mathrm{ik}}\left|\right)}^2+\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{K=1}{\overset{N}{\mathrm{\Σ}}}\underset{i\≠j}{\overset{M}{\underset{j=1}{\mathrm{\Σ}}}}\underset{i=1}{\overset{n}{\mathrm{\Σ}}}(1-X_{\mathrm{ij}})a_{\mathrm{iK}}{a}_{\mathrm{jK}}$

Wherein, a _iKbe the i-th row K row channel usage, a _jKfor jth row K row channel usage;

(4) to the selection operational module of different I D grouping: selection operation is on the basis of ideal adaptation degree assessment in all groupings being based upon multiple different I D, in assumed group, certain individual B ' fitness is f (B '), then certain individual B ' is P by the probability selected _s=f (B ')/Σ f (B '), ideal adaptation degree is larger, higher by the probability selected, and ideal adaptation degree is less, lower by select probability, so just can determine that those individualities are selected;

(5) channel assignment module, it is according to genetic operator, determines final Channel distribution mode, and wherein, genetic operator comprises the crossover operator of improvement and the mutation operator of improvement, the crossover probability P of the crossover operator wherein improved _cchange formula is:

$P_c=\left\{\begin{array}{cc}{P}_{c2}-\frac{(P_{c2}-P_{c1})\×(f_{\mathrm{avg}}-f_i)}{(f_{\mathrm{avg}}-f_{\min })}& f_i\≤f_{\mathrm{avg}}\\ P_{c3}-\frac{(P_{c3}-P_{c2})\×(f_{\max }-f_i)}{(f_{\max }-f_{\mathrm{avg}})}& f_i>f_{\mathrm{avg}}\end{array}\right.$

Wherein, f _avgfor the whole individual average fitness value of contemporary population, f _maxfor maximum adaptation angle value individual in population, f _minfor minimum fitness value individual in population, f _ibe that in 2 intersection individualities, fitness value is larger, P _c1, P _c2, P _c3represent the value of three different phase crossover probabilities respectively, P _c1represent f _icrossover probability when=0, P _c2represent f _i=f _avgtime crossover probability, P _c3represent f _i=f _maxtime crossover probability;

The mutation probability P of the mutation operator improved _mformula as follows:

$P_m=\left\{\begin{array}{cc}{P}_{m1}& f_i\≤f_{\mathrm{avg}}\\ P_{m2}-\frac{(P_{m2}-P_{m1})\×(f_{\max }-f_i)}{(f_{\max }-f_{\mathrm{avg}})}& f_i>f_{\mathrm{avg}}\end{array}\right.$

Wherein f _ifor chromosome fitness value, f _avgfor population average fitness value, P _m1, P _m2represent the numerical value of two different phase mutation probability respectively, P _m1represent f _i≤ f _avgtime mutation probability, P _m2represent f _i> f _avgtime mutation probability;

End condition, when algorithm evolution is to regulation algebraically, the solution that the individual B ' that fitness value f (B ') is corresponding finally finds for algorithm.