CN104581918B - Satellite layer-span combined optimization power distribution method based on non-cooperative game - Google Patents

Abstract

The invention relates to a satellite cross-layer joint optimization power allocation method based on static non-cooperative game. The method comprises: synchronous broadband satellite physical layer statistics of real-time information of each user; considering that the satellite is a power-limited system, according to the QoS needs of different services, the power allocation problem is modeled as a multi-user static non-cooperative game model; according to the solution: Allocate power for different services and adopt corresponding coding methods for different services at the application layer; repeat the above steps regularly. The cross-layer optimized power allocation method provided by the present invention can enable the satellite network control center to model the service power allocation problem as a multi-service static non-cooperative game model according to the channel state feedback information provided by the physical layer and considering the QoS requirements of different services. At the same time, according to different cross-layer solutions, corresponding coding methods are adopted for different services in the application layer, which can comprehensively improve the communication quality of users in the mobile communication system and improve system performance.

Description

A method of satellite cross-layer joint optimization power allocation based on non-cooperative game

technical field

The invention relates to the technical field of space communication, in particular to a non-cooperative game-based satellite cross-layer joint optimization power allocation method.

Background technique

With the commercialization of terrestrial cellular 4G technology gradually matured, research on 5G technology is also in full swing. The mobile satellite communication system introduced in order to solve the coverage problem of the terrestrial cellular mobile communication system has become a research hotspot accordingly. Among them, the integrated mobile communication system obtained by integrating the satellite mobile communication system and the ground mobile communication system is considered to be an important part of the next generation communication network.

For the synchronous satellite communication system, since the satellite is in a synchronous satellite orbit far from the ground, the RTT (Round-Trip Time, round-trip delay) of the data packet is as long as 500ms, making it difficult for the satellite to obtain CSI (Channel State Information, channel state information) in time Resource management, so in order to provide effective QoS (Quality of Service, Quality of Service) guarantee for on-board forwarding services, mature and complex resource management and scheduling schemes on the ground are not suitable for direct use on-board. Due to the technical conditions of existing satellite architecture design and practical factors such as satellite-ground distance and space interference, the broadband satellite communication system is actually a power-limited system (here specifically refers to the on-board architecture), so when designing a satellite resource management solution Further research on power control and distribution is required.

From the perspective of the function realization of radio resource management, the radio resource management system includes core functions such as satellite MAC (Media Access Control, media access control layer) protocol, access control protocol, link layer bandwidth allocation, and physical layer resource allocation. The traditional satellite radio resource management system is designed in layers, and often only optimizes each layer independently, seldom considering the joint optimization performance of the entire network. In order to further make full use of the scarce power and bandwidth resources on the star, it is necessary to adopt a cross-layer resource management design, and perform joint optimization under the conditions of specific service type QoS guarantee and system resource constraints. While extending the guarantee, it maximizes the satisfaction of other services and optimizes the overall performance of the broadband satellite mobile communication system.

Contents of the invention

In view of the above defects, the present invention provides a cross-layer power optimization method based on multi-user static non-cooperative game, which can maximize the satisfaction of other services while ensuring the QoS of services with strict requirements on time delay.

A cross-layer power optimization method based on multi-user static non-cooperative game, specifically including:

S1: Under the control of the satellite network control center, the synchronous broadband satellite physical layer counts the real-time information of each business and uploads it to the satellite network control center;

S2: The satellite network control center defines the utility function of the network based on the physical layer statistical information and the QoS requirements of different services, and models the service power allocation problem as a multi-service static non-cooperative game model;

S3: According to the revenue vector of the model, allocate the overall useful power to different beams according to the service and system requirements, and adopt corresponding coding methods for different services at the application layer;

S4: Repeat steps S1-S3 according to a predetermined cycle.

Further, the real-time information of each service counted by the physical layer of the synchronous broadband satellite in step S1 includes: coding and modulation mode, transmission power, signal-to-noise ratio, available bandwidth and bit error rate, etc.;

Further, the real-time information in step S1 is used for the joint optimization scheme design of the upper layer and the physical layer under the control of the satellite network control center, so as to maximize the satisfaction of multi-service;

Further, the satellite network control center in step S1 analyzes the real-time information to obtain satellite channel state information;

Further, step S2 also includes the satellite network control center estimating the previous multi-service power allocation scheme according to the satellite channel state information, and determining the QoS satisfaction degree of different services;

Further, the satellite network control center in step S2 establishes a multi-service static non-cooperative game model according to the feedback satellite channel state information and the QoS requirements of different services under the constraint condition of on-board transmission power, specifically including:

S21: Assuming that there are N satellite services competing for the useful power capacity of the satellite link in the system, each service i has a minimum guaranteed power P ₁ and a variable enhanced power P ₂ ;

S22: In the model, each business iteratively updates its own strategy. In each iteration, the current user chooses a strategy that maximizes its own utility function, while the strategies of other users remain unchanged;

S23: Perform a limited number of iterations according to the model and satisfy the power constraint condition until a converged optimal solution is obtained or the number of iterations is exhausted.

Further, in each iteration, the user selects the optimal strategy according to the channel state of the terminal and the satellite channel information fed back by the satellite control center;

Further, the channel and service of the satellite are constant within a time window, so signaling resources can be saved;

Further, the utility function is:

in, The power allocation request selected for the i-th service, in bit/s; The maximum utility function obtained for the i-th service based on the power allocation requests of other services;

Further, the overall useful power allocation of the satellite based on the model optimization solution described in step S3 follows the following principles:

The useful power allocated to the business of each terminal is not lower than the minimum guaranteed power, and not higher than the maximum useful power value that can be allocated by the beam;

According to the optimal solution of the model and the service QoS requirements, priority is given to terminals with good channel conditions or services with high QoS requirements to allocate enhanced power;

For delay-insensitive services, the coding method is adaptively adjusted at the application layer to reduce the bit error rate.

Further, the overall useful power allocated to the service of each terminal;

Further, the satellite channel status and services are considered constant during the predetermined period in step S4; the channel state information is no longer counted within the predetermined period, so as to save signaling overhead.

The invention provides a cross-layer optimization satellite power allocation method based on a static non-cooperative game model. Under the control of the satellite network control center, the satellite counts the real-time information of each terminal service and uploads it to the network control center. The network control center defines the effect function based on channel information and service QoS requirements, and models the power allocation problem as a static non-cooperative game model, and solves it for a limited number of iterations. According to the revenue vector of the model, the satellite network control center allocates useful power to each beam and different types of services, and at the same time adaptively adjusts the coding method at the application layer for delay-insensitive services to reduce the bit error rate. Therefore, the improved cross-layer optimization method of the present invention can optimize power allocation in satellites based on service QoS requirements and channel status, improve communication quality, and improve overall communication performance.

Description of drawings

Fig. 1 is the flowchart of the cross-layer optimization power allocation method based on the static non-cooperative game model of the present invention;

Fig. 2 is a schematic diagram of distribution of satellites and users according to an embodiment of the present invention.

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 is a method flowchart of an embodiment of the present invention, which specifically includes the following steps:

Step S1: Under the control of the satellite network control center, the synchronous broadband satellite physical layer counts the real-time information of each service, and uploads it to the satellite network control center;

Step S2: The satellite network control center defines the utility function of the network based on the physical layer statistical information and the QoS requirements of different services, and models the service power allocation problem as a multi-service static non-cooperative game model;

Step S3: According to the revenue vector of the model, the overall useful power is allocated to different beams according to the business and system requirements, and corresponding coding methods are adopted for different businesses at the application layer;

Step S4: Repeat the above steps according to a predetermined period.

Through the above steps, under the control of the network control center, the satellite defines the effect function based on the channel information and service QoS requirements, establishes a static non-cooperative game model, and performs a limited number of iterations to obtain the optimal solution, and uses the minimum guaranteed power based on the effective power And the enhanced power is allocated to different types of services under each beam, and the coding method is adaptively adjusted at the application layer for delay-insensitive services to comprehensively reduce the bit error rate, improve the communication quality of the satellite system, and improve the overall communication performance.

Preferably, the real-time information of each service collected by the satellite physical layer includes: coding and modulation scheme, transmission power, signal-to-noise ratio, available bandwidth, bit error rate, and the like. The real-time information is used in the joint optimization scheme design of the upper layer and the physical layer under the control of the network control center to maximize the satisfaction of multiple services.

The establishment of the static non-cooperative game model is based on the known channel state information and the QoS requirements of each service. In the network control center, the useful power allocation method allocated to each service is modeled and optimized to achieve the best overall efficiency.

The establishment process of the static non-cooperative game model includes:

S21: Assuming that there are N satellite services competing for the useful power capacity of the satellite link in the system, each service i has a minimum guaranteed power P ₁ and a variable enhanced power P ₂ ;

S22: In the model, each business iteratively updates its own strategy. In each iteration, the current user chooses a strategy that maximizes its own utility function, while the strategies of other users remain unchanged;

S23: Carry out a limited number of iterations according to the model and satisfy the power constraint condition until a converged optimal solution is obtained or the number of iterations is exhausted;

Here we define that a service i in each beam selects a power allocation request x _req,i (bit/s).

As the variable strategy competition capacity C, the variable strategy set for each service i is defined as S _i . The variable strategy space of the game is S=S ₁ ×S ₂ ×…×S _N , so a variable strategy combination is an N-dimensional vector:

The goal of every business is to maximize its own utility It determines the request expectation obtained by adjusting x _req,i for each business. The solution for is defined as follows:

In order to establish a link between the considered business model and the utility function, a QoS priority weight parameter Ω _i >0 is introduced here, defined as:

Among them, q _j is the weight assigned to each type of business, h _i (j) is the percentage of type j business of user i, so the utility function can be further expressed as:

Preferably, in each iteration, the user selects the optimal strategy according to the channel state of the terminal and the satellite channel information fed back by the satellite control center;

Preferably, the power allocation process for each beam and service according to the static non-cooperative game model includes:

S31: Allocate useful power not lower than the minimum guaranteed power to the service of each terminal;

S32: According to the optimal solution of the model and the service QoS requirements, preferentially allocate enhanced useful power to terminals with good channel conditions or services with high QoS requirements;

S33: For delay-insensitive services, adaptively adjust the encoding method at the application layer to reduce the bit error rate.

Preferably, the satellite channel state and service are considered constant during the predetermined period in step S4; no statistics on channel state information are made during the predetermined period in step S4, so as to save signaling overhead.

An example of the present invention is given below.

This embodiment considers the distribution of service power at the edge of a satellite beam, as shown in FIG. 2 .

In the predetermined period, according to the statistical real-time information and the QoS requirements of each business, a static non-cooperative game model is established and iterated several times to obtain an optimal solution. And allocate useful power to different types of services of each beam according to the optimal solution.

For beam edge users A and B shown in the figure, the channel conditions are very poor. Considering that the service types of the two are different, user A is a mobile terminal, and the service type is voice service; user B is a streaming media service such as video. The two traffic types have different QoS weights when modeling accordingly. After obtaining the optimal solution, user A needs to allocate enhanced power on the basis of the minimum guaranteed power to ensure voice service quality; user B needs to adjust the coding method at the application layer to reduce the bit error rate and improve communication performance.

The above embodiments are only used to illustrate the present invention, but not to limit the present invention. Those of ordinary skill in the relevant technical field can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, all Equivalent technical solutions also belong to the category of the present invention, and the scope of patent protection of the present invention should be defined by the claims.

Claims (10)

1. A satellite cross-layer joint optimization power distribution method based on a static non-cooperative game is characterized by specifically comprising the following steps:

s1: under the control of a satellite network control center, a synchronous broadband satellite physical layer counts real-time information of each service and uploads the real-time information to the satellite network control center;

s2: the satellite network control center defines a utility function of the network based on the physical layer statistical information and QoS requirements of different services, and models a service power distribution problem into a multi-service static non-cooperative game model, wherein the static non-cooperative game model establishing process comprises the following steps:

n satellite services in the system are supposed to compete for the useful power capacity of a satellite link, and each service i has the lowest guaranteed power and variable enhanced power;

in the model, each service iteratively updates its own strategy, and in each iteration, the current user selects a strategy capable of maximizing its own utility function, while the strategies of other users remain unchanged;

carrying out finite iteration under the condition of meeting the power constraint condition according to the model until a convergence optimal solution is obtained or the iteration times are used up;

s3: distributing the whole useful power to different beams according to the service and system requirements according to the income vector of the model, and adopting corresponding coding modes aiming at different services at an application layer;

s4: steps S1 to S3 are repeated according to a predetermined cycle.

2. The method of claim 1, wherein the step S1 of synchronizing the real-time information of each service of the broadband satellite physical layer statistics comprises: code modulation mode, transmission power, signal-to-noise ratio, available bandwidth and bit error rate.

3. The method of claim 1, wherein the real-time information of step S1 is used for joint optimization scheme design of upper layer and physical layer under control of a network control center to achieve maximum satisfaction of multiple services.

4. The method of claim 1, wherein step S1 the network control center analyzes the real-time information to obtain satellite channel state information.

5. The method of claim 1, wherein in each iteration, the user selects the optimal strategy according to the channel state of the terminal and satellite channel information fed back by the satellite control center.

6. The method of claim 1, wherein the channel and traffic of the satellite are constant over a time window.

7. The method of claim 1, wherein the utility function is:

wherein,selecting a power distribution request for the ith service, wherein the unit is bit/s;for the ith service based on othersThe maximum utility function obtained by the power allocation request of the service.

8. The method of claim 1, wherein the method for allocating the satellite total useful power based on the model optimization solution in step S3 is as follows:

the useful power allocated to the service of each terminal is not lower than the minimum guaranteed power and not higher than the maximum useful power allocable by the wave beam;

preferentially distributing enhanced power for a terminal with good channel condition or a service with high QoS requirement according to the optimal solution of the model and the QoS requirement of the service;

aiming at the time delay insensitive service, the coding mode is adaptively adjusted at the application layer to reduce the error rate.

9. The method of claim 8, wherein the total useful power allocated for the traffic of each terminal is not higher than the maximum useful power value that a beam can be allocated.

10. The method of claim 1, wherein the satellite channel state and traffic are considered constant for a predetermined period in step S4; and not counting the channel state information in the preset period.