CN105979542A - SDN (Software Defined Network) based WiFi shunting mechanism in 5G heterogeneous network - Google Patents

Abstract

The invention relates to an SDN (Software Defined Network) based WiFi shunting mechanism in 5G heterogeneous network, and belongs to the technical field of mobile communication. The mechanism specifically comprises that when terminal UE sends a service request frame, an SDN controller provides an available AP list for the requesting UE; an SDN monitoring module feeds state information of network equipment and the requesting UE back to a shunt decision-making module of an SDN, channel load judgment is carried out on all APs in the list at first in the shunt decision-making module, the APs which are not overloaded are classified into a group, a channel load value of each AP in the list is calculated, and then the throughout Pw and the effective data receiving rate R(t) of all APs in the list are calculated in sequence; and finally, the channel load value, the throughout Pw and the effective data receiving rate R(t) of each AP are assigned with corresponding weight factors, a utility function QoC of the requesting UE and each AP in the second list is acquired through multiplication and accumulation, the AP with the maximum QoC value is enabled to act as an optimal shunting target, and the requesting UE is notified to access to the target AP so as to realize shunting. The shunting mechanism provided by the invention can realize seamless integration of 5G and WiFi, thereby dealing with explosive increasing of mobile data traffic, effectively solving a severe data overload problem of the cellular network, and improving the user experience under a heterogeneous network environment.

Description

SDN-based WiFi shunting mechanism in 5G heterogeneous network

Technical Field

The invention belongs to the technical field of mobile communication, and relates to a WiFi shunting mechanism based on SDN in a 5G heterogeneous network.

Background

In recent years, the rapid development of mobile terminals and mobile internet services brings huge data bearing pressure to cellular networks, and the explosive increase of mobile data traffic causes the cellular networks to face serious data overload problems. Heterogeneous networks (Het-Nets) have become a key network architecture for solving the requirements of high capacity and wide coverage in a future 5G network, and access network control of multi-layer deployment in the heterogeneous networks is realized by Radio Access Technologies (RATs). Typical deployments are those covering very dense small cells such as Picos, Femtos, relay nodes, WiFi access points, etc. within a macro cell. Due to the characteristics of high bandwidth, low cost, convenient deployment, and small interference with the cellular network, the WiFi wireless network has been taken as an important measure for operators to deal with the rapidly increasing data traffic.

In a future 5G communication system, along with effective fusion of a mobile network and a WiFi network, perception difference of a terminal user to the mobile network and the WiFi network may disappear, the user can seamlessly and smoothly transit from the mobile network to the WiFi network, and the seamless fusion of the 5G and the WiFi network enables the terminal to be automatically connected with an optimal network and gives no perception experience to the user. The effective WiFi shunting mechanism can better promote seamless fusion of 5G and WiFi, through a proper WiFi shunting scheme, the overload of a cellular network is reduced, meanwhile, more reliable service quality is provided for users, network resources of each access point are fully utilized, and the resource utilization rate of the whole network is maximized.

The Access Network Discovery and Selection Function (ANDSF) is a new element proposed in stage 3GPP R8, and has data management and control functions, which are capable of providing assistance data required for discovering a user's access network and for making an access network selection in response to a user's request for making an access network selection. And the user selects the most appropriate access network according to the information provided by the ANDSF to realize data service distribution. However, the ADNSF is a policy based on a server in the domain of the mobile network operator, the network discovery and selection policy is pre-configured by the mobile network operator, and such static configuration does not reflect (display) the current state of the access network, cannot dynamically obtain the network performance index in real time, and cannot effectively provide the user with the best access network.

SDN (Software Defined Network) is a new feasible technology to control a future 5G communication Network through intellectualization, and a control layer of a router is separated from a data layer by using an OpenFlow protocol and implemented in a Software manner. The SDN controller can carry out overall monitoring on the network, grasp the latest dynamics of network equipment in real time and have certain visibility on realizing data distribution. Based on SDN technology, data distribution is easier through centralized coordination and reliable control of various network infrastructures.

Based on the technical scheme, the invention provides a prospective WiFi shunting mechanism based on SDN in a 5G heterogeneous network according to the development trend of mobile communication technology.

Disclosure of Invention

In view of this, the present invention provides a WiFi offloading mechanism based on SDN in a 5G heterogeneous network, which can implement seamless convergence between 5G and WiFi in the future, cope with explosive growth of mobile data traffic, effectively solve the problem of serious data overload faced by a cellular network, and improve user experience in a heterogeneous network environment.

In order to achieve the purpose, the invention provides the following technical scheme:

a WiFi offloading mechanism based on SDN in a 5G heterogeneous network is realized based on SDN architecture in a network where an MBS and a plurality of WiFiAps are converged and coexisted, and specifically comprises the following steps:

s1: the SDN controller monitors the state information of the network equipment in real time, when the mobile terminal UE sends a service request frame, the SDN controller checks whether available WiFiAps exist around the request terminal UE, and WiFi roaming is allowed;

s2: if the allowed WiFi roaming exists, namely an accessible AP exists, the SDN controller lists APs accessible to the request terminal UE as a group, and generates a list NAP-list 1; otherwise, the requesting UE stays on the current cellular network;

s3: the SDN acquires network state information of Aps in the NAP-list1 list, sequentially judges whether channel loads of the Aps in the list exceed a set load threshold theta according to channel load conditions, and if the channel loads exceed the set load threshold theta, the request terminal UE still stays in the current network; otherwise, listing a group of non-overloaded Aps lists, namely NAP-list 2;

s4: sequentially calculating the throughput P of each AP in the NAP-list2 according to the APs network state information acquired by the SDN controller_wAnd carrying out normalization processing on the data;

s5: obtaining a data transmission rate T (t) between the mobile terminal and each AP in a NAP-list2 list at a time t according to a Shannon formula, obtaining an effective data receiving rate R (t) ═ T (t) (1-Z (t)), wherein Z (t) is a transmission rate error rate, and standardizing R (t);

s6: constructing a utility function: load of channel, throughput P to AP respectively_wAnd the effective data receiving rate R (t) is endowed with corresponding weight factors to obtain utility function QoC values between the requesting UE and each AP in the NAP-list2 list;

s7: and taking the AP with the maximum QoC value as a shunting target, executing shunting by the SDN controller according to a shunting algorithm result, and accessing the request UE to the optimal target AP to realize shunting.

Further, the SDN controller includes two functional modules: the monitoring module and the shunt decision module can monitor network state information globally, communicate with network equipment through an OpenFlow protocol, and realize centralized management of network resources.

Further, the monitoring module can monitor the network states of the mobile terminal UE, its neighboring BSs and the WiFi AP in real time, and feed back the acquired network information resources and the service request of the mobile terminal to the offloading decision module.

Further, after receiving the network state information from the monitoring module, the offloading decision module first performs channel Load judgment on available Aps (NAP-list 1) adjacent to the mobile terminal UE sending the service request frame, and lists the Aps not overloaded as a group, i.e., NAP-list2, where the channel Load is defined by IEEE 802.11K and is obtained by formula (1):

${\mathrm{Load}}_i=\frac{Channel\_Busy\_Fraction}{Measurement\_Duration}---\left(1\right)$

wherein, Measurement _ Duration is a Channel Measurement time, and Channel _ Busy _ Fraction is a Channel Busy time.

Further, for each AP in the list NAP-list2 in the split decision module, according to the publicEquation (2) calculates the throughput P of each AP in the list in turn_wAnd normalizing the data:

$P_{w_i}=\frac{{\mathrm{np}}_{tr}{(1-p_{tr})}^{n-1}L}{1+\{{\Σ}_{j=1}^n{(1-p_{tr})}^{n-1}(\frac{L}{C_j}+T_0)+\[1-{(1-p_{tr})}^n-{\mathrm{np}}_{tr}{(1-p_{tr})}^{n-1}\]T_c\}}---\left(2\right)$

wherein n represents the number of users in the AP; l represents the payload length of the user; c_jRepresenting the rate at which the user is communicating; p is a radical of_trThe transmission probability of the user in the WiFi in the correct transmission state is obtained; t is_oRepresenting the transmission overhead (including detecting the channel, authenticating, and user attribution, etc.); t is_cThe number of time slots required for each collision time.

Further, the offloading decision module calculates the corresponding effective data receiving rate according to the link information of the requesting terminal UE and each AP in the list NAP-list2, and standardizes the effective data receiving rate, where the formula is as follows:

1)representing terminal UE and available AP_iOf a hypothetical linkThe bandwidth at time t isHz, obtaining the terminal UE and AP according to the Shannon formula_iThe data transmission rate of (c) is:

$T_i^{UE}\left(t\right)=b_i^{UE}\left(t\right)(1+\frac{s_i^{UE}\left(t\right)}{n_i^{UE}\left(t\right)})---\left(3\right)$

wherein,is a linkThe power of the received signal at the time t,is additive white gaussian noise;

2) due to noise interference, the transmitted data will have certain errorsRepresentative linkThe error rate at time t is calculated as:

wherein,

3) the link is obtained by the formulas (3) and (4)Effective data receiving rate at time t

$R_i^{UE}\left(t\right)=T_i^{UE}\left(t\right)(1-e_i^{UE}(t\left)\right)---\left(5\right).$

Further, a channel Load parameter Load and a throughput parameter P for the AP are respectively set_wAnd the effective data receiving rate parameter R (t) is given a respective weighting factor α, β, gamma, and 0<α<1，0<β<1，0<γ<1, α + β + γ ═ 1, and defines the requesting terminal UE and the available APs in the list NAP-list2_iThe utility function QoC of AP_iLoad_iThroughput ofEffective data receiving rateAnd multiplying and accumulating by corresponding weight factors to obtain:

${\mathrm{QoC}}_i=\mathrm{\&alpha;}\frac{1}{{\mathrm{Load}}_i}+{\mathrm{\&beta;P}}_{wi}+{\mathrm{\&gamma;R}}_i^{UE}\left(t\right)---\left(6\right).$

further, the AP with the largest QoC value is taken as the optimal offloading target, and the requesting UE is notified to access the target AP, thereby implementing offloading.

The invention has the beneficial effects that: based on an SDN platform architecture, the method and the system comprehensively acquire the network condition information of each access point by utilizing the advantages of the SDN, such as global real-time monitoring of the state information of the network equipment and excellent network management capability, comprehensively consider the channel load condition and throughput of the access AP and the effective data receiving rate of the request terminal UE, obtain the maximum utility function between the UE and the available AP, and select the optimal access AP for the UE as a shunting target. The flow distribution mechanism effectively relieves the data flow pressure borne by the cellular network, prevents network congestion and improves the user experience in the heterogeneous network environment.

Drawings

In order to make the object, technical scheme and beneficial effect of the invention more clear, the invention provides the following drawings for explanation:

fig. 1 is a schematic diagram of a 5G network scenario based on SDN;

fig. 2 is a flow chart of the shunting mechanism according to the present invention.

Detailed Description

Preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

Fig. 1 is a schematic diagram of a 5G network scenario based on SDN suitable for use in an embodiment of the present invention. In order to meet the requirements of future networks and services, an effective shunting method is realized on a 5G network architecture based on an SDN. In the scene, in the coverage range of the cellular network, a plurality of access points WiFi Aps capable of being covered again are arranged, the coverage radius of the access points WiFi Aps is far smaller than that of the cellular network, the SDN is located at the control layer, the data layer is separated from the control layer through an OpenFlow protocol, and the data layer is communicated with network equipment through the OpenFlow protocol. The SDN control mainly comprises two parts: a monitoring module and a shunt decision module. The monitoring module monitors the network information of the data layer equipment in real time, feeds the network information back to the distribution decision module, and selects the optimal access AP for the user through the decision execution of a distribution algorithm, so that the data flow overload problem of the cellular network can be effectively relieved while the user experience under the heterogeneous network environment is improved.

Fig. 2 is a schematic flow chart of the shunting mechanism according to the present invention, and as shown in the figure, the method specifically includes the following steps:

1. in the coverage range of one MBS, the number of {1,2, …, I } WiFi Aps capable of being re-covered exists, when a mobile terminal UE sends a service request frame, an SDN controller checks whether the WiFi Aps available around the requesting UE exist according to the network equipment state information obtained by monitoring, and if so, the SDN controller lists the WiFi Aps as a group of available Aps lists (namely NAP-list 1); otherwise, stay on the current cellular network.

2. After the SDN controller acquires the network state information of the Aps in the NAP-list1 list, sequentially judging whether the channel load of the Aps in the list exceeds a set threshold theta according to a channel load calculation formula, and if so, requesting the terminal UE to stay in the current connection network; in contrast, a set of non-overloaded Aps lists (i.e., NAP-list2) is listed, and the channel load formula is:

${\mathrm{Load}}_i=\frac{Channel\_Busy\_Fraction}{Measurement\_Duration}$

the Channel Load is defined according to IEEE 802.11k, where Measurement _ Duration is the Channel Measurement time and Channel _ Busy _ Fraction is the Channel Busy time.

3. After step 2 is completed, respectively calculating the throughput of each Aps in the list and the effective data receiving rate between the Aps and the requesting UE, and performing normalization processing on the throughput and the effective data receiving rate. The calculation formulas are respectively as follows:

AP_ithroughput of (2):

$P_{w_i}=\frac{{\mathrm{np}}_{tr}{(1-p_{tr})}^{n-1}L}{1+\{{\&Sigma;}_{j=1}^n{(1-p_{tr})}^{n-1}(\frac{L}{C_j}+T_0)+\&lsqb;1-{(1-p_{tr})}^n-{\mathrm{np}}_{tr}{(1-p_{tr})}^{n-1}\&rsqb;T_c\}}$

wherein n represents the number of users in the AP; l represents the payload length of the user; c_jIndicating the speed at which the user is communicatingRate; p is a radical of_trThe transmission probability of the user in the WiFi in the correct transmission state is obtained; t is_oRepresenting the transmission overhead (including detecting the channel, authenticating, and user attribution, etc.); t is_cThe number of time slots required for each collision time.

AP_iEffective data reception rate with the requesting UE:

S1：representing terminal UE and available AP_iOf a hypothetical linkThe bandwidth at time t isHz, obtaining the terminal UE and AP according to the Shannon formula_iThe data transmission rate of (c) is:

$T_i^{UE}\left(t\right)=b_i^{UE}\left(t\right)(1+\frac{s_i^{UE}}{n_i^{UE}\left(t\right)})$

wherein,is a linkThe power of the received signal at the time t,is additive white gaussian noise.

S2: due to noise interference, the transmitted data will have certain errorsRepresentative linkThe error rate at time t is calculated as:

wherein,

s3: the link is obtained from S1 and S2Effective data receiving rate at time t

$R_i^{UE}\left(t\right)=T_i^{UE}\left(t\right)(1-e_i^{UE}-(t\left)\right)$

Normalization processing formula:

$G_w=\frac{P_w-P_{min}}{P_{\max }-P_{min}}$

4. channel Load parameter Load, throughput parameter P for AP_wAnd the effective data receiving rate parameter R (t) is given a respective weighting factor α, β, gamma, and 0<α<1，0<β<1，0<γ<1, α + β + γ ═ 1, and defines requesting terminal UE and available APs_iThe utility function QoC of AP_iLoad_iThroughput ofEffective data receiving rateAnd multiplying and accumulating by corresponding weight factors to obtain:

${\mathrm{QoC}}_i=\mathrm{\&alpha;}\frac{1}{{\mathrm{Load}}_i}+{\mathrm{\&beta;P}}_{wi}+{\mathrm{\&gamma;R}}_i^{UE}\left(t\right)$

5. and taking the AP with the maximum QoC value as an optimal shunting target, and informing the requesting UE to access the target AP to realize shunting.

Finally, it is noted that the above-mentioned preferred embodiments illustrate rather than limit the invention, and that, although the invention has been described in detail with reference to the above-mentioned preferred embodiments, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the scope of the invention as defined by the appended claims.

Claims (8)

1. An SDN-based WiFi offloading mechanism in a 5G heterogeneous network, characterized in that: in a network where an MBS and a plurality of WiFi aps coexist in a converged manner, a WiFi offloading mechanism is implemented based on an SDN architecture, which specifically includes the following steps:

s1: the SDN controller monitors the state information of the network equipment in real time, when the mobile terminal UE sends a service request frame, the SDN controller checks whether available WiFiAps exist around the request terminal UE, and WiFi roaming is allowed;

s2: if the allowed WiFi roaming exists, namely an accessible AP exists, the SDN controller lists APs accessible to the request terminal UE as a group, and generates a list NAP-list 1; otherwise, the requesting UE stays on the current cellular network;

s3: the SDN acquires network state information of Aps in the NAP-list1 list, sequentially judges whether channel loads of the Aps in the list exceed a set load threshold theta according to channel load conditions, and if the channel loads exceed the set load threshold theta, the request terminal UE still stays in the current network; otherwise, listing a group of non-overloaded Aps lists, namely NAP-list 2;

s4: sequentially calculating the throughput P of each AP in the NAP-list2 according to the APs network state information acquired by the SDN controller_wAnd carrying out normalization processing on the data;

s5: obtaining a data transmission rate T (t) between the mobile terminal and each AP in a NAP-list2 list at a time t according to a Shannon formula, obtaining an effective data receiving rate R (t) ═ T (t) (1-Z (t)), wherein Z (t) is a transmission rate error rate, and standardizing R (t);

s6: constructing a utility function: load of channel, throughput P to AP respectively_wAnd the effective data receiving rate R (t) is endowed with corresponding weight factors to obtain utility function QoC values between the requesting UE and each AP in the NAP-list2 list;

s7: and taking the AP with the maximum QoC value as a shunting target, executing shunting by the SDN controller according to a shunting algorithm result, and accessing the request UE to the optimal target AP to realize shunting.

2. The SDN-based WiFi offload mechanism in a 5G heterogeneous network of claim 1, wherein: the SDN controller comprises two functional modules: the monitoring module and the shunt decision module can monitor network state information globally, communicate with network equipment through an OpenFlow protocol, and realize centralized management of network resources.

3. The SDN-based WiFi offload mechanism in a 5G heterogeneous network of claim 2, wherein: the monitoring module can monitor the network states of the mobile terminal UE, the BSs adjacent to the mobile terminal UE and the WiFi AP in real time, and feeds back the acquired network information resources and the service request of the mobile terminal to the distribution decision module.

4. The SDN-based WiFi offloading mechanism in a 5G heterogeneous network of claim 3, wherein: after receiving the network state information from the monitoring module, the offloading decision module first determines channel Load of neighboring available Aps, that is, NAP-list1, of the UE that sends the service request frame, and lists the Aps that are not overloaded as a group, that is, NAP-list2, where the channel Load is defined by IEEE 802.11K and is obtained by formula (1):

${\mathrm{Load}}_i=\frac{Channel\_Busy\_Fraction}{Measurement\_Duration}---\left(1\right)$

wherein, Measurement _ Duration is a Channel Measurement time, and Channel _ Busy _ Fraction is a Channel Busy time.

5. The SDN-based WiFi offloading mechanism in a 5G heterogeneous network of claim 4, wherein:in the shunting decision module, for each AP in the list NAP-list2, sequentially calculating the throughput P of each AP in the list according to the formula (2)_wAnd normalizing the data:

$P_{w_i}=\frac{{\mathrm{np}}_{tr}{\left(1-p_{tr}\right)}^{n-1}L}{1+\left\{{\mathrm{\&Sigma;}}_{j=1}^n{\left(1-p_{tr}\right)}^{n-1}\left(\frac{L}{C_j}+T_0\right)+\&lsqb;1-{\left(1-p_{tr}\right)}^n-{\mathrm{np}}_{tr}{\left(1-p_{tr}\right)}^{n-1}\&rsqb;T_c\right\}}---\left(2\right)$

wherein n represents the number of users in the AP; l represents the payload length of the user; c_jRepresenting the rate at which the user is communicating; p is a radical of_trThe transmission probability of the user in the WiFi in the correct transmission state is obtained; t is_oRepresenting the transmission overhead (including detecting the channel, authenticating, and user attribution, etc.); t is_cThe number of time slots required for each collision time.

6. The SDN-based WiFi offloading mechanism in a 5G heterogeneous network of claim 5, wherein: the offloading decision module respectively calculates the corresponding effective data receiving rate according to the request terminal UE and the link information of each AP in the list NAP-list2, and standardizes the effective data receiving rate, where the formula is as follows:

1)representing terminal UE and available AP_iOf a hypothetical linkThe bandwidth at time t isHz, obtaining the terminal UE and AP according to the Shannon formula_iThe data transmission rate of (c) is:

$T_i^{UE}\left(t\right)=b_i^{UE}\left(t\right)\left(1+\frac{s_i^{UE}\left(t\right)}{n_i^{UE}\left(t\right)}\right)---\left(3\right)$

wherein,is a linkThe power of the received signal at the time t,is additive white gaussian noise;

2) due to noise interference, the transmitted data will have certain errorsRepresentative linkThe error rate at time t is calculated as:

wherein,

3) the link is obtained by the formulas (3) and (4)Effective data receiving rate at time t

$R_i^{UE}\left(t\right)=T_i^{UE}\left(t\right)\left(1-e_i^{UE}\left(t\right)\right)---\left(5\right).$

7. The SDN-based WiFi offloading mechanism in a 5G heterogeneous network of claim 6, wherein: channel Load parameter Load, throughput to AP respectivelyParameter P_wAnd the effective data receiving rate parameter R (t) is given a respective weighting factor α, β, gamma, and 0<α<1，0<β<1，0<γ<1, α + β + γ ═ 1, and defines the requesting terminal UE and the available APs in the list NAP-list2_iThe utility function QoC of AP_iLoad_iThroughput ofEffective data receiving rateAnd multiplying and accumulating by corresponding weight factors to obtain:

${\mathrm{QoC}}_i=\mathrm{\&alpha;}\frac{1}{{\mathrm{Load}}_i}+{\mathrm{\&beta;P}}_{w_i}+{\mathrm{\&gamma;R}}_i^{UE}\left(t\right)---\left(6\right).$

8. the SDN-based WiFi offload mechanism in a 5G heterogeneous network of claim 7, wherein: and taking the AP with the maximum QoC value as an optimal shunting target, and informing the requesting UE to access the target AP to realize shunting.