CN102711179B - Multi-service dynamic bandwidth distribution method under heterogeneous wireless network environment - Google Patents

Abstract

The invention belongs to wireless communication technology field, disclose the Multi-service dynamic bandwidth distribution method under a kind of heterogeneous wireless network environment, comprising: S1, when receiving the access request of business, collecting the multiple domain information of each network; S2, classifies to the business of request access, is divided into variable bit rate VBR business and fixed bit rate CBR business, according to the characteristic of VBR business, formulates VBR service bandwidth and distributes planning strategy; S3, chooses corresponding analytic hierarchy process (AHP) discrimination matrix, and application level analytic approach calculates the adaptability factor of each accessible network adaptive; S4, for heterogeneous networks state, application utility function sets up the effectiveness fair bandwidth allocation strategy introducing the adaptability factor, and multiple network is for asking access service distribution strip wide.What the invention solves that user fairness guarantee and systematic function promote takes into account problem, makes full use of the overall performance of isomerism with elevator system of network, effectively improves resource utilization ratio.

Description

Multi-service Dynamic Bandwidth Allocation Method in Heterogeneous Wireless Network Environment

technical field

The invention belongs to the technical field of wireless communication, and in particular relates to a multi-service dynamic bandwidth allocation method in a heterogeneous wireless network environment.

Background technique

According to the different business characteristics and user requirements, a variety of wireless access technologies have emerged. The vigorous development of various wireless communication technologies makes the future wireless communication system present a heterogeneous pattern. With the evolution of information and communication technology, the wireless access network is developing towards the trend of high speed, broadband and ubiquitous. Faced with so many wireless access networks with different characteristics, how to design a unified wireless communication network resource management strategy to fully accommodate them, truly fundamentally realize flexible and optimized wireless resource allocation, and provide users with seamless, high-quality rich Service experience is a hot topic in current heterogeneous network research.

Different wireless access networks have relatively large differences in network coverage, service capabilities, and usage costs, and service types, rates, and QoS (Quality of Service) requirements are also different. If a user is in the intersection of multiple wireless access networks at the same time Within the coverage area, it is necessary to select and use a network access combination based on certain policies or standards. At the same time, in hotspot coverage areas, due to the increase in the number of users and business load, resource shortages or unbalanced loads between networks may be caused, resulting in a significant increase in business blocking and dropped calls, which will not only affect the user's business experience, but also It will cause a lot of resource consumption, how to achieve load balancing between networks under heavy load, how to ensure user fairness caused by uneven network coverage in the case of heterogeneous networks, and how to ensure efficient bandwidth utilization of different wireless access networks Efficiency, increasing network capacity and avoiding network overload, optimizing resource allocation in joint radio resource management, realizing bandwidth allocation in a heterogeneous wireless access environment, efficient use of wireless resources, reasonable resolution of user fairness, and timely satisfaction of user business needs , which poses a great challenge to wireless resource allocation in heterogeneous networks.

Contents of the invention

(1) Technical problems to be solved

In view of the defects and deficiencies in existing resource allocation schemes, the purpose of the present invention is to provide a resource allocation scheme in a heterogeneous wireless network environment, so that multi-network bandwidth allocation of user requested services can reflect user fairness and network performance improvement Two aspects, that is, to realize efficient utilization of heterogeneous network resources under the premise of guaranteeing user QoS fairness requirements.

(2) Technical solutions

In order to achieve the above object, the present invention provides a multi-service dynamic bandwidth allocation method in a heterogeneous wireless network environment, comprising the following steps:

S1. When a service access request is received, collect multi-domain information of each network, and classify the multi-domain information into multi-attribute domain information;

S2. Classify the services requesting access, and divide them into variable bit rate VBR services and fixed bit rate CBR services. According to the characteristics of VBR services, formulate a VBR service bandwidth allocation planning strategy. For VBR services, use VBR service bandwidth allocation planning A policy calculates the total bandwidth allocated to it by multiple networks;

S3. Applying the network state dynamic threshold method, selecting a corresponding AHP discriminant matrix for different network states, and applying the AHP obtained according to the multi-domain information to obtain adaptability factors for adapting each accessible network;

S4. According to different network states, a utility function is applied to establish a utility fair bandwidth allocation policy that introduces an adaptive factor, and multiple networks allocate bandwidth for requesting access services.

Preferably, the multi-domain information includes the set of accessible networks requesting access to the service, and the quality of service (QoS) parameters of the networks in the set of accessible networks. The QoS parameters include parameters reflecting user fairness and parameters used to reflect network performance Parameters for improving requirements. Correspondingly, classifying the information of multi-attribute domains is specifically to divide QoS parameters into parameters for reflecting user fairness and parameters for reflecting network performance improvement requirements.

Preferably, the user fairness of the heterogeneous wireless network is reflected in that when a relatively large-scale coverage access network is consumed in a relatively small-scale area where multiple networks overlap For access users covered by the network, user fairness is reduced; the parameters used to reflect network performance improvement requirements include parameters used to reflect network throughput and blocking rate.

Preferably, in step S2, the services requesting access are divided according to the service type and the range of bandwidth requirements, and the services whose bandwidth requirements are fixed bit rate are divided into CBR services, and the services whose bandwidth requirements are variable bit rate are divided into VBR services. business.

Preferably, the VBR service bandwidth allocation planning strategy is: count and average all existing VBR service bandwidths in the area where the service requesting access is located, and obtain the average value as the total number of newly arrived VBR services allocated by multiple networks. The target value for the bandwidth.

Preferably, in step S3, an AHP discrimination matrix corresponding to different network states is established according to the level of network load.

Preferably, between step S2 and step S3, according to the coverage characteristics of heterogeneous wireless networks, multiple wireless access networks covering an area are divided into two categories, the first category is the wireless access network with the smallest area covering the area network, the second type is other wireless access networks covering the area; in step S3, the dynamic threshold method of the network status is specifically: according to the load level of the second type of wireless access network, dynamically adjust the threshold for judging the first type The threshold of the load state of the radio access network is used to ensure user fairness requirements.

Preferably, the utility function is used to measure the quantitative relationship between the bandwidth allocated by the network to the newly arrived service and the performance change of the access network. In step S4, the utility function is first applied to quantify the bandwidth allocated to the service, and then the bandwidth is allocated using the utility fairness The policy divides the network state into two situations: full load and low load. In the case of low load, the bandwidth ratio is calculated according to the following principles: VBR service bandwidth allocation target is the same as the first arriving VBR service allocation target, and CBR service allocates bandwidth according to bandwidth requirements; In the case of full load, the bandwidth ratio is calculated according to the following principles: the service being served reduces its own bandwidth according to the principle of fairness to provide for newly arrived services, and if the reduced bandwidth does not meet the minimum bit rate requirement, the newly arrived services will be blocked.

Preferably, according to whether the radio access network corresponding to the area can accept new services, the network status is divided into two situations: full load and low load.

(3) Beneficial effects

Compared with the prior art, the resource allocation method proposed by the present invention divides services into multiple sub-flows for transmission on different networks, which not only realizes the balance between user QoS fairness and network system performance in heterogeneous scenarios, but also fully Utilizes the transmission capability of the wireless access network in the heterogeneous wireless environment, efficiently utilizes the network resources, and optimizes the performance of the system.

Description of drawings

FIG. 1 is a schematic diagram of a scenario of multi-network coverage areas in a heterogeneous environment according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of a multi-discriminant matrix AHP model structure in a heterogeneous wireless access scenario according to an embodiment of the present invention;

FIG. 3 is a schematic flowchart of a resource allocation method in a heterogeneous wireless scenario according to an embodiment of the present invention;

Fig. 4 compares the scheme of the embodiment of the present invention with the load balancing strategy, and draws a comparison diagram of user blocking rates in different regions during the system running time;

Fig. 5 is a comparison chart of the total blocking rate of the heterogeneous network system during the system running time obtained by comparing the solution of the embodiment of the present invention with the load balancing strategy.

Detailed ways

The specific implementation manners of the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

Fig. 1 shows the scene of the heterogeneous wireless network in this embodiment, in the framework model of network structure applied in this embodiment, the geographic area considered is completely covered by Wireless Metropolitan Area Network (WMAN), 3G wireless access network base station and wireless local area network (WLAN) hotspots only cover part of the area. Area A1 is only covered by wireless metropolitan area network, area A2 is covered by wireless metropolitan area network (WMAN) and 3G network, and area A3 is covered by three kinds of networks. Wireless Metropolitan Area Network (WMAN) is based on the IEEE 802.16e protocol, with a spectrum range of 2-66GHz, affected by factors such as transmission bandwidth and modulation and coding methods, and supports a business data reduction rate of 10-100M. The 3G wireless access network is based on the access technology of TD-SCDMA (Time Division-Synchronous Code Division Multiple Access), and the system capacity is 2Mbps. The WLAN network is based on the IEEE 802.11 protocol, and the maximum bandwidth supports 11Mbps. In this embodiment, it is assumed that the mobile terminal can access the wireless access network corresponding to the area where it is located, and on the basis of power control, the transmission rate of the wireless access network in the coverage area is evenly distributed.

The multi-network cooperative transmission resource allocation method mentioned in this embodiment comprehensively considers the role of parameters that reflect user QoS fairness and network system performance improvement, and applies dynamic thresholds and multi-discriminant matrix analytic hierarchy process according to different network load levels Integrate the two aspects to make multi-factor decision-making, so as to obtain a resource allocation plan that balances the two aspects.

In this embodiment, the service arrival time of users in areas A1, A2, and A3 obeys the Poisson distribution, and the average arrival rate is shown in Table 1, and the service time of users accessing the network follows a negative exponential distribution.

Table 1 Arrival rate in different regions

As shown in Figure 3, the resource allocation method under the heterogeneous wireless network of the embodiment of the present invention specifically includes the following steps:

S1. The user terminal sends an access request, triggers the collection of multi-domain information of each wireless access network, and classifies the multi-domain information into information of multi-attribute domains. Collect context information that reflects two aspects of user fairness and wireless network performance improvement. In this embodiment, user QoS fairness is mainly reflected in the access fairness of users at the edge of a large-scale access network. When a large-scale wireless access network is consumed in a small area where multiple networks overlap Access users who only access the network for coverage in a large area will reduce user fairness; the demand for wireless network performance improvement is reflected in the improvement of network-side performance parameters such as throughput and blocking rate of multi-network heterogeneous systems.

In this embodiment, the contents of the following four parameters are specifically collected:

Parameters used to reflect the demand for wireless network performance improvement: the system capacity C of the wireless access network and the current available capacity C′ of the wireless network. The system capacity C of the wireless access network represents the intrinsic carrying capacity of the network and the potential of the network to carry service loads, and C' represents the current capacity of the wireless access network to carry the load level.

Parameters for user fairness: the coverage area S of the wireless access network and the user traffic density ρ corresponding to the area covered by the wireless access network. In order to ensure the fairness of users at the edge of the large-scale coverage network, when the business in the small-scale multi-network coverage area allocates bandwidth from multiple networks, the bandwidth of the small-area coverage network should be used first; similarly, when the small-scale multi-network coverage When the business density of an area increases, the proportion of bandwidth allocated from the large-scale access network in this area should be reduced to reduce the excessive consumption of large-scale network bandwidth in a small-scale area.

S2. Summarize and classify the requested wireless services into variable bit rate VBR services and fixed bit rate CBR services, and formulate VBR service bandwidth allocation planning strategies according to the characteristics of the VBR services;

Among them, according to the type of service and the range of bandwidth requirements, the services requesting access (referred to as request services) are divided, the services requiring bandwidth at a fixed bit rate are divided into CBR services, and the services requiring bandwidth at a variable bit rate are divided into For VBR business. For example, the voice service requires a fixed-rate bandwidth, and some background services such as Internet access have more flexible bandwidth supply. In this embodiment, the fixed bandwidth requirement of the CBR service is 256Kbps, and the bandwidth requirement of the VBR service is 256Kbps to 512Kbps.

For VBR services, it is first necessary to calculate the total bandwidth allocated to VBR service requests by different networks according to the VBR service bandwidth allocation planning strategy. The VBR service bandwidth allocation planning policy counts and averages all the VBR service bandwidths in the area where the requested service is located, as a new To reach the target value of the total bandwidth allocated by multiple networks for VBR services, specifically:

For the first VBR service arriving in the kth area, the maximum bandwidth requirement of the first VBR service is taken as the target value B _{N+1, k} , namely:

${\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; B</span>}}_{\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; B</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}^{\mathrm{<span\; class="notranslate">\; VBR</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>$

Among them, N represents the current total number of services in the area, that is, N+1 can be expressed as the newly arrived services in the current area, N ₂ is the number of VBR services corresponding to the k-th area, and B _{N+1, k} is the newly arrived VBR services The allocated bandwidth target of The maximum bandwidth required by the VBR service. B _{N+1, k} is jointly provided by multiple networks covering this area. Assuming that there are m heterogeneous wireless access networks in this area, then:

${\mathrm{<span\; class="notranslate">\; B</span>}}_{\mathrm{<span\; class="notranslate">\; B</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; b</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; k</span>}}\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1,2</span>}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span>$

In the formula, b _{i, N+1, k} is the bandwidth allocated by the i-th wireless access network for the new VBR service.

There is a VBR service situation for the kth area:

${\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ></span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; B</span>}}_{\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; no</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; B</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; /</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1,2</span>}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 3</span>}<span\; class="notranslate">\; )</span>$

Based on this, compared with the existing VBR traffic in the kth area, the newly arrived VBR traffic fairly obtains the target value of bandwidth allocation.

S3, applying the network state dynamic threshold method, selecting the corresponding AHP discriminant matrix according to the load level of different networks, and applying the multi-discriminant matrix AHP to calculate the adaptability factor parameters for each accessible wireless network;

Multi-discriminant matrix analytic hierarchy process:

According to the above discussion of user fairness in this embodiment, when applying the AHP to calculate the adaptability factors for each wireless access network, the two factors of network system performance improvement and user fairness should be considered comprehensively, and based on these The transformation of factors is adjusted dynamically in real time.

According to the two types of network parameters collected in step S1 of this embodiment, the AHP algorithm model (as shown in Figure 2) is established, and the four parameters are combined to finally calculate the adaptability to the characteristics of each wireless access network factor parameter. The judgment matrix of the scheme layer and the criterion layer is mainly based on the operator's experience data or business strategy, and can be set artificially. The discrimination matrix between the criterion layer and the target layer is mainly set based on the importance of the fairness parameters and network system performance parameters relative to the adaptability factors.

The network load level is divided into three states: high, medium, and low. In the AHP model, the discrimination matrix W of the criterion layer relative to the target layer is set as W ^h , W ^m , and W ^l , respectively. corresponds to a network state.

Further analysis shows that when the network is under low load, all arriving new services, including CBR services and VBR services, can obtain bandwidth according to the maximum bandwidth requirement. Therefore, when the bandwidth level of the heterogeneous network system is not high, priority should be given to satisfying System performance improvement requirements; and when the heterogeneous network system is under high load, the excessive consumption of the large-scale coverage network in the small-scale multi-network coverage area will inevitably cause high blocking rates for edge users. When loaded, the importance of parameters related to fairness increases.

Therefore, by solving the discriminant matrix of the criterion layer and the target layer, the importance of the four network parameters of the criterion layer relative to the adaptability factor is obtained as w=[w _c ′, w _s , w _ρ , w _c ]. And corresponding to the three network states, when the load level increases, the importance of the fairness parameter increases, reflecting the decrease in the importance of the network performance improvement parameter, that is, the importance of the solution needs to meet:

${\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; c</span>}}^{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{{\mathrm{<span\; class="notranslate">\; c</span>}}^{<span\; class="notranslate">\; \&prime;</span>}}^{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; \&Greater\; Equal;</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; c</span>}}^{\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{{\mathrm{<span\; class="notranslate">\; c</span>}}^{<span\; class="notranslate">\; \&prime;</span>}}^{\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; \&Greater\; Equal;</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; c</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{{\mathrm{<span\; class="notranslate">\; c</span>}}^{<span\; class="notranslate">\; \&prime;</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 4</span>}<span\; class="notranslate">\; )</span>$

${\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}}^{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; \&rho;</span>}}^{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; \&le;</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}}^{\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; \&rho;</span>}}^{\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; \&le;</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; \&rho;</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 5</span>}<span\; class="notranslate">\; )</span>$

Dynamic threshold method:

The threshold is a key factor to determine the discontinuous state of the network, that is, the network is divided into three states by using the threshold, namely, high, medium and low. Determining the threshold is a key step in dividing these three states. This embodiment introduces the concept of a dynamic threshold to further ensure user fairness.

According to the coverage characteristics of heterogeneous networks, various wireless access networks covering an area are divided into two categories. The first type is the wireless access network with the smallest area covering the area, and the second is other wireless access networks covering the area. network. As shown in FIG. 1 , for the users in area A3, the WiMax network and the cellular network belong to the second type of network, while the WLAN network belongs to the first type of network.

In this embodiment, it is considered that the kth area is covered by m types of wireless access networks, wherein the nth network is the network with the smallest coverage area covering the area, that is, the first type of network. is the threshold applied by the second type of network to judge the network status; ${\mathrm{\&theta;}}_1^h,{\mathrm{\&theta;}}_2^h({\mathrm{\&theta;}}_1^h\&le;{\mathrm{\&theta;}}_2^h),$ ${\mathrm{\&theta;}}_1^m,{\mathrm{\&theta;}}_2^m({\mathrm{\&theta;}}_1^m\&le;{\mathrm{\&theta;}}_2^m),$ ${\mathrm{\&theta;}}_1^l,{\mathrm{\&theta;}}_2^l({\mathrm{\&theta;}}_1^l\&le;{\mathrm{\&theta;}}_2^l)$ is the threshold used by the first type of network (that is, network n) to judge the network status.

The process of applying the dynamic threshold method to select the discriminant matrix W in the multi-discriminant matrix analytic hierarchy process applies the following steps:

(1), according to the network load degree of the second type network, determine the threshold combination for judging the load state of the first type network, there are three situations:

$1\&le;\&ForAll;i\&le;\mathrm{mandi}\&NotEqual;\mathrm{no},{\mathrm{\&theta;}}_i\&Greater\; Equal;{\mathrm{\&theta;}}_2^{\mathrm{classl}},$ That is, when all the second-type networks are in a high-load state, the combination of the first-type load state discrimination thresholds is selected as

$1\&le;\&ForAll;i\&le;\mathrm{mandi}\&NotEqual;\mathrm{no},\&Exists;{\mathrm{\&theta;}}_i\&le;{\mathrm{\&theta;}}_1^{\mathrm{claal}},$ That is, when the second type of network is in a low-load state, the combination of thresholds for selecting the first type of load state is

When the above two conditions are not satisfied, the threshold combination for selecting the first type of network to judge the load status is ${\mathrm{\&theta;}}_1^m,{\mathrm{\&theta;}}_2^m({\mathrm{\&theta;}}_1^m\&le;{\mathrm{\&theta;}}_2^m).$

(2) Apply the threshold combination selected in step (1) and the first type of network load level to determine the discrimination matrix for the application. For example, when the selection result of the first step is , the selection process is as follows: When, W=W ^h ; When, W=W ^m ; , W=W ^l .

Based on the previous definition of the discriminant matrix, the discriminant matrix W ^h , W ^m , and W ^l correspond to the high, medium, and low load levels of the network respectively. When the load is high, the first type of network should select the threshold to make it easier to select the discrimination matrix corresponding to the high load level, so that it is easier to enter the state of guaranteeing edge users, so ${\mathrm{\&theta;}}_1^h,{\mathrm{\&theta;}}_2^h({\mathrm{\&theta;}}_1^h\&le;{\mathrm{\&theta;}}_2^h),$ ${\mathrm{\&theta;}}_1^m,{\mathrm{\&theta;}}_2^m({\mathrm{\&theta;}}_1^m\&le;{\mathrm{\&theta;}}_2^m),$ ${\mathrm{\&theta;}}_1^l,{\mathrm{\&theta;}}_2^l({\mathrm{\&theta;}}_1^l\&le;{\mathrm{\&theta;}}_2^l)$ Should meet:

${\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; \&le;</span>{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; \&le;</span>{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 6</span>}<span\; class="notranslate">\; )</span>$

${\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; \&le;</span>{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; \&le;</span>{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 7</span>}<span\; class="notranslate">\; )</span>$

S4, applying a utility function to establish a utility-fair bandwidth allocation policy that introduces an adaptive factor according to different network states, and multiple networks allocate bandwidth for requesting access services.

In this step, a utility function is first introduced to measure the quantitative relationship between the bandwidth allocated by the network to newly arrived services and the performance change of the access network:

The utility function is a concept in economics. In economics, it is used to express the quantitative relationship between the utility obtained by consumers in consumption and the combination of commodities consumed. In theory, it can approximate the utility function of each person through a series of psychological tests. It is used to measure the degree of satisfaction consumers get from consuming a given combination of goods.

In the application of bandwidth allocation strategy, the utility function is defined as a monotonically increasing function, which is used to represent the mapping relationship between the change of network performance and the bandwidth allocated by the network to a certain service. The utility function is used to roughly estimate the impact on network performance when the system allocates bandwidth for a newly arrived connection, so as to quantify the allocated bandwidth of the business. The expressions that define the bandwidth allocated by the network and the utility obtained by the network are:

U(P)＝ωlog(1+α·P/ε) (8)

In the formula, P is the bandwidth allocated by the wireless access network for new arrival services. Due to the characteristics of heterogeneous networks, an adaptability factor ε needs to be introduced to make the utility function adapt to different networks. At the same time, ω and α are constant factors, which are convenient The utility function is adjusted on the basis of the adaptability factor, and U(P) is the utility value obtained by allocating bandwidth P to the network.

In the utility fair bandwidth allocation strategy, according to whether the wireless access network system corresponding to the area can accept new services, the network status is divided into two situations: full load and low load. The remaining bandwidth of the access network system cannot meet the needs of newly arrived services. The specific strategies in the two cases are:

At low load:

In the low-load state, the VBR service allocates bandwidth according to the results of the bandwidth allocation planning strategy (according to the VBR bandwidth allocation planning strategy, when the network is under low load, the VBR service bandwidth allocation target is the same as the first arriving VBR service allocation target, that is, the maximum demand allocation), the CBR service allocates bandwidth according to the bandwidth demand.

Therefore, for CBR and VBR services, when a new service arrives, the bandwidth allocation satisfies:

$\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; \&le;</span><span\; class="notranslate">\; \&ForAll;</span>{\mathrm{<span\; class="notranslate">\; i</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; i</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; u</span>}}_{{\mathrm{<span\; class="notranslate">\; i</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; b</span>}}_{{\mathrm{<span\; class="notranslate">\; i</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; u</span>}}_{{\mathrm{<span\; class="notranslate">\; i</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; b</span>}}_{{\mathrm{<span\; class="notranslate">\; i</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 9</span>}<span\; class="notranslate">\; )</span>$

Therefore, it is guaranteed that different networks in the k-th area have equal utility values for allocating bandwidth for newly arriving services. By solving equations (14), (9), (8), (2), and (1) or (3), the proportion of bandwidth allocated by different networks for new arrival services can be obtained:

p _i,N+1 ,k=bi _,N+1,k /B _N+1,k (10)

because is a constant factor of the k-th area of the i-th network and is artificially set and known. Therefore, for the newly arriving traffic in the k-th area, the adaptability factor of the i-th network Determines the bandwidth allocation ratio of the multi-network to newly arrived services.

At full load:

For the full load state, the remaining bandwidth of the wireless access network system corresponding to the area cannot meet the needs of newly arrived services. Therefore, the VBR services currently serving in the area reduce the bandwidth to the newly arrived services according to the principle of fairness. If the VBR services reduce the bandwidth If the minimum bit rate requirement is not met, the newly arrived business will be blocked. Specifically follow the following three steps to calculate:

1. According to the utility fair bandwidth allocation strategy, calculate the bandwidth allocated by different networks for the new arrival service. The bandwidth of b _i,N+1,k .

2. In each wireless access network, all serving VBR services will reduce the bandwidth fairly according to formula (11)(12), β _{i, j, k} means that for the i-th access network in the k-th area Bandwidth reduced by j VBR services.

$\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; \&le;</span><span\; class="notranslate">\; \&ForAll;</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; \&le;</span><span\; class="notranslate">\; \&ForAll;</span>{\mathrm{<span\; class="notranslate">\; j</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; j</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; \&le;</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; j</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; /</span>{\mathrm{<span\; class="notranslate">\; b</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; j</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; j</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; /</span>{\mathrm{<span\; class="notranslate">\; b</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; j</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 11</span>}<span\; class="notranslate">\; )</span>$

${\mathrm{<span\; class="notranslate">\; b</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; j</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1,2</span>}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 12</span>}<span\; class="notranslate">\; )</span>$

3. If the VBR services of all access networks in the k-th area satisfy the formula (13), then admit the services, otherwise, block the services.

$\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; \&le;</span><span\; class="notranslate">\; \&ForAll;</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; \&le;</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ,</span>{{\mathrm{<span\; class="notranslate">\; B</span>}}^{<span\; class="notranslate">\; \&prime;</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; B</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; -</span>\underset{\mathrm{<span\; class="notranslate">\; j</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&Greater\; Equal;</span>{\mathrm{<span\; class="notranslate">\; B</span>}}_{\mathrm{<span\; class="notranslate">\; min</span>}}^{\mathrm{<span\; class="notranslate">\; VBR</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 14</span>}<span\; class="notranslate">\; )</span>$

Thus, through different processing mechanisms under light network load and full network load, the calculation of multi-network bandwidth bearing for newly arrived services is satisfied, and network resources are effectively utilized.

Fig. 4 and Fig. 5 are comparison results of the solution and the load balancing strategy of the embodiment of the present invention. Among them, under the load balancing strategy, the multi-network bandwidth allocation ratio of the newly arrived service is a fixed value, and its value is the bandwidth ratio of different access network systems. When the network is fully loaded, the VBR service also reduces the bandwidth to meet the bandwidth requirements of newly arrived services.

FIG. 4 is a comparison diagram of user blocking rates in different areas during the system running time obtained by comparing the scheme of the embodiment of the present invention with the load balancing strategy. Through comparison, it can be found that the solution proposed in the embodiment can effectively reduce the blocking rate in all areas; and the fairness of wide coverage of network edge users is significantly guaranteed.

Fig. 5 is a comparison chart of the total blocking rate of the heterogeneous network system during the system running time obtained by comparing the solution of the embodiment of the present invention with the load balancing strategy. Through comparison, it can be found that the solution proposed in the embodiment utilizes heterogeneous network system resources more effectively, and the system blocking rate is significantly reduced.

It can be seen from the comparison that the resource allocation method proposed by the present invention has better system performance.

The foregoing is only an embodiment of the present invention. It should be pointed out that for those of ordinary skill in the art, some improvements and modifications can be made without departing from the technical principle of the present invention. It should be regarded as the protection scope of the present invention.

Claims (6)

1. A multi-service dynamic bandwidth allocation method under a heterogeneous wireless network environment, characterized in that, comprising the following steps: S1. When a service access request is received, collect multi-domain information of each network, and classify the multi-domain information into multi-attribute domain information; S2. Classify the services requesting access, and divide them into variable bit rate VBR services and fixed bit rate CBR services. According to the characteristics of VBR services, formulate a VBR service bandwidth allocation planning strategy. For VBR services, use VBR service bandwidth allocation planning A policy calculates the total bandwidth allocated to it by multiple networks; S3. Applying the network state dynamic threshold method, selecting a corresponding AHP discriminant matrix for different network states, and applying the AHP obtained according to the multi-domain information to obtain adaptability factors for adapting each accessible network; S4, for different network states, apply a utility function to establish a utility fair bandwidth allocation strategy that introduces an adaptive factor, and multiple networks allocate bandwidth for requesting access services; Wherein, the utility function is used to measure the quantitative relationship between the bandwidth allocated by the network to the newly arrived service and the performance change of the access network. In step S4, the utility function is first applied to quantify the allocated bandwidth of the service, and then the utility fair bandwidth allocation strategy is used Divide the network state into two cases of full load and low load. In the case of low load, the bandwidth ratio is calculated according to the following principles: the bandwidth allocation target of the VBR service is the same as that of the first arriving VBR service, and the bandwidth of the CBR service is allocated according to the bandwidth demand. Different networks have the same utility value for allocating bandwidth for newly arriving services; under full load, the bandwidth ratio is calculated according to the following principles: the service being served reduces its own bandwidth to provide for newly arriving services according to the principle of fairness, if the bandwidth is not satisfied after reducing the bandwidth If the minimum bit rate is required, the newly arrived traffic is blocked; Wherein, in step S3, according to the height of the network load level, establish the AHP discriminant matrix corresponding to different network states; Between step S2 and step S3, according to the coverage characteristics of heterogeneous wireless networks, multiple wireless access networks covering an area are divided into two types, the first type is the wireless access network with the smallest area covering the area, and the second type is the wireless access network covering the area. The second type is other wireless access networks covering the area; in step S3, the network state dynamic threshold method is specifically: according to the load level of the second type of wireless access network, dynamically adjust the wireless access network used to judge the first type The threshold of network load status to ensure user fairness requirements. 2. The method according to claim 1, wherein the multi-domain information includes the set of accessible networks requesting access to the service, and the quality of service (QoS) parameters of networks in the set of accessible networks, QoS parameter It includes parameters used to reflect user fairness and parameters used to reflect network performance improvement requirements. Correspondingly, the information classification of multi-attribute domains is specifically divided into parameters used to reflect user fairness and parameters used to reflect user fairness. Parameters for network performance improvement requirements. 3. The method according to claim 2, wherein the user fairness is embodied in that when a relatively large coverage access network is consumed in a relatively small area covered by overlapping networks, for For access users only covered by a large-scale access network in the surrounding area, user fairness is reduced; the parameters used to reflect the network performance improvement requirements include parameters used to reflect the network throughput and blocking rate. 4. The method according to claim 1, characterized in that, in step S2, according to the business type and the scope of bandwidth requirements, the services requested for access are divided, and the bandwidth requirements are divided into CBR services for fixed bit rate services, and the The bandwidth requirement is the business of variable bit rate is divided into VBR business. 5. The method according to claim 1, characterized in that, the VBR service bandwidth allocation planning strategy is: statistics of all existing VBR service bandwidths in the area where the service requested to access is located and averaging, and the obtained average value is used as The target value of the total bandwidth allocated by multiple networks for newly arrived VBR traffic. 6. The method according to any one of claims 1-5, characterized in that, in the utility fair bandwidth allocation strategy, the network status is divided into full and full according to whether the wireless access network corresponding to the area can accept new services load and low load.