CN-117222015-B - Online SVC multicast method based on unmanned aerial vehicle relay in NOMA network

Abstract

The invention provides an online SVC multicast method based on unmanned aerial vehicle relay in a NOMA network, and aims to maximize video receiving quality at the edge of a base station. In the method, a visual graph model is firstly constructed to describe the coupling between unmanned aerial vehicle deployment and multicast group association. Based on the model, the video reception quality maximization problem is modeled as a nonlinear integer programming problem based on cliques clique, and is decoupled from the two sub-problems of drone-multicast group association and sub-channel allocation. The former is converted into a maximum weight group problem with vertex number limitation, and an improved maximum weight group algorithm based on branch delimitation is adopted to determine the unmanned aerial vehicle-multicast group association mode. For the latter, a rule matching policy is employed, an optimal resource allocation policy is obtained with less computational cost. Simulation results show that the method is superior to the existing reference scheme in the aspects of signal-to-noise ratio PSRN of the aggregate peak value, spectrum utilization rate, adaptability and the like.

Inventors

• BAI GUANGWEI
• WANG YUANYI
• SHEN HANG
• TONG ZIYUAN
• WANG TIANJING

Assignees

• 南京工业大学

Dates

Publication Date

20260505

Application Date

20230828

Claims (3)

1. 1. In an online SVC video multicast scene of unmanned aerial vehicle relay, a plurality of unmanned aerial vehicles are deployed as relays on the coverage edge of a macro base station, wherein three different types of links are: the ground base station-unmanned plane B2U link is used for transmitting the SVC base layer to the unmanned plane by the base station; a ground base station-to-ground equipment B2D link for the base station to send the SVC enhancement layer to the user; the unmanned aerial vehicle to ground equipment U2D link is used for transmitting the SVC base layer to a user; Users requesting the same video stream belong to a multicast group; the mobile edge computing MEC controller determines the deployment position of the unmanned aerial vehicle and the association mode of the unmanned aerial vehicle and the multicast group by accessing global information, and allocates spectrum resources for each multicast group; The online SVC multicast method is characterized by comprising the following steps: step 1) constructing a visual graph model for describing the coupling between the placement of the unmanned aerial vehicle and the association of the multicast group; Step 2) modeling a video reception quality maximization problem as a nonlinear integer programming problem p 1 based on a clique clique based on a visual graph model; step 3) decoupling the problem p 1 into an unmanned aerial vehicle placement-multicast group association sub-problem and a sub-channel allocation sub-problem; step 4) solves two sub-problems: Step 4.1) converting the unmanned aerial vehicle placement-multicast group association sub-problem into a maximum weight group problem that the undirected graph in the visual graph model has vertex number limitation, and then adopting an improved maximum weight group algorithm based on branch delimitation to determine an unmanned aerial vehicle placement-multicast group association mode; Step 4.2) adopting a rule matching strategy to solve the problem of sub-channel allocation, and obtaining an optimal resource allocation strategy with small calculation cost; In the step 1), a visualized graph model is constructed to describe the coupling between different decision variables, and support is provided for unmanned aerial vehicle deployment and multicast group association: obtaining a set of unmanned aerial vehicle candidate positions by using a clustering algorithm for users in all multicast groups Wherein Is an index of the projection position on the X-Y plane, Representing a candidate position of the unmanned aerial vehicle on the X-Y axis; Order the Representing an undirected graph, each vertex of the undirected graph corresponds to a candidate decision for placement of a drone associated with the multicast group and must satisfy Then, the vertex set is expressed as , Wherein, the For a set of multicast group indexes, Is associated with A height index corresponding to the position; representing the maximum signal gain from the macro base station m to the coverage edge of the unmanned aerial vehicle; P s represents the transmission power of the unmanned aerial vehicle, and p m represents the transmission power of the base station; When any two are located at And When the drones of (a) are all associated with multicast group n, and (3) with The corresponding constraint is re-expressed as (A) Indicating the maximum channel gain from the second drone to the first drone in coverage, Representing the minimum channel gain from the macro base station m to the first unmanned aerial vehicle coverage area; a unmanned aerial vehicle hovering over the plane position index j can only select a unique height index k, corresponding to (B) If and only if (A) or (B) is true, the first unmanned aerial vehicle corresponds to the vertex Vertex corresponding to second unmanned plane With one edge in between, whereby the set of edges is expressed as , Representative of A clique Clique, clique Clique is a subset of vertices in the undirected graph, where there must be a connection for any two vertices; Each of the clusters Are mapped into a set containing "drone placement-multicast group association" decision variables; In the step 2), the problem of maximizing video quality is converted into a problem of spectrum division based on clustering, namely, a cluster is found to determine unmanned plane placement and multicast group association, and the number of sub-channels of each multicast group is determined, specifically: Definition q j,k,n is used to determine if vertex v j,k,n is in the selected clique Inner part , The 0-1 variable u 1,n,i 、u 2,n,i represents whether the user i in the multicast group n receives the base layer and the enhancement layer, 0 represents the receiving, and 1 represents the not receiving; The aggregate signal-to-noise ratio PSNR of the received video for multicast group n is shown as one for sub-channels b n and Is a function of (i.e.) , Then pair In a multicast group application function The video quality maximization problem P1 is then modeled as , s.t. (a) (b) (c) (d) (e) (f) (g) (h) (i) Q j,k,n corresponds to the determination of the first drone and its vertex v j,k,n , Corresponds to a second unmanned aerial vehicle and its vertex Is judged by (1); Representing a minimum bit rate that the intra-group ground device supports normal decoding when the multicast group n requests the base layer; representing a minimum bit rate that the intra-group ground device supports normal decoding when the multicast group n requests an enhancement layer; representing a user Decoding an achievable rate of the drone signal from location l j,k ; Is shown in the user Decoding the reachable rate of the base station m signal; Constraint (a) represents the point pair , ) There are no edges and the two vertices do not belong to the same cluster; constraint (b) represents that one unmanned aerial vehicle is allowed to serve a plurality of multicast groups, constraint (b) comprises a sign function For calculating clusters The number of unmanned aerial vehicles required to be launched must be equal to or less than the number of available unmanned aerial vehicles When placed at When no multicast group is associated with the drone, Otherwise, 1; in the constraint (c) and the constraint (d), Represents a constant large enough to ensure , In the case of u 1,n,i =1、u 1,n,i =0, user i can or cannot receive and decode the base layer, respectively; Constraint (e) and constraint (f) Is a constant large enough to ensure , In the case of u 1,n,i =1, u 2,n,i =1 or u 2,n,i =0 indicates that user i can or cannot receive and decode the enhancement layer, respectively; Constraint (g) ensures that the sum of the number of sub-channels allocated to each multicast group does not exceed the total number B held by the base station; In the step 3), the problem is that Is decoupled into a drone-multicast group association sub-problem and a sub-channel allocation sub-problem; In the step 4.1), solving the unmanned aerial vehicle-multicast group association sub-problem: The weights of the vertices represent the contribution of vertex v j,k,n to the video quality improvement, the weights of the vertices being defined as , The unmanned plane placement-multicast group association sub-problem is converted into a maximum weight group problem P1.1 with vertex number constraint in search, and is described as , s.t.(a), (b), (c),(d),(e),(f),(h),(i) Searching the maximum weight group by adopting an improved maximum weight group algorithm based on branch delimitation to solve the problem P1.1; Inputs to the algorithm include: : is a maximum weight group that has been found in the (c), The set of candidate vertices currently being processed, To and from the vertex The set of all vertices connected, the initial phase, And The air is set to be in the air, Is that All vertices of (a); Before the algorithm is executed in advance of the execution of the algorithm, The method is characterized in that the method comprises the steps of inputting the maximum weight group into a delimiting function to obtain the weight value of the maximum weight group, marking the weight value as t as the upper bound of the maximum weight group in a subgraph, wherein the algorithm comprises the following steps: If it is The upper weight bound of (2) is not greater than the current maximum weight group Then at On termination of the recursive search, return On the contrary, from Selecting the vertex with the largest weight ; Obtaining a dough The number of unmanned aerial vehicles required if the number exceeds Skipping searches at that point, otherwise adding that point to ; Then at Selection and selection All vertices of the join line as new candidate set In the following Performing the algorithm recursively; If from Middle return bolus Weight of (2) is greater than Then update And from Is removed from ; Continuing to search for eligible vertices until Empty, return ; And (3) with The corresponding optimal decision is , Representing the vertex with the greatest weight In the ball An inner part; in the step 4.2), the sub-channel allocation sub-problem is solved: From the output of step 4.1) The subchannel allocation molecular problem P1.2 is modeled as , s.t. , User' s The minimum rates required to receive the base layer and enhancement layer are expressed as , And , Is shown in The channel gain from the unmanned aerial vehicle to the ground equipment i in the coverage range; representing base station m to ground equipment Channel gain of (a); Is shown in The channel gain from the unmanned aerial vehicle to the ground equipment i in the coverage range; When the minimum receiving rate requirements of both base layer and enhancement layer decoding are met, then no further increase of subchannels is required, the largest subchannels required for multicast group n are denoted as , Representing a user The rate required by the base layer is received, Representing a user Receiving a rate required by the enhancement layer; Considering the influence of the multicast group number N and the subchannel number B on the calculation complexity, adopting a rule as follows to match a strategy: a) When (when) When the PSNR value of each multicast group is higher than the PSNR value of the other multicast groups, the PSNR value of each multicast group is higher than the PSNR value of each multicast group, and the PSNR value of each multicast group is higher than the PSNR value of each multicast group; b) When (when) When all multicast group users can receive the base layer and enhancement layer videos, multicast group n is allocated A sub-channel; c) When (when) When a modified knapsack algorithm is used to determine the spectral division: N multicast groups are considered N types of items, each type having B items, which need to be placed in a backpack of capacity B; In the nth category The weight of the individual articles is It profit is obtained from Determining; for the remaining b sub-channels, the maximum PSNR of the first n multicast groups is marked as If (3) Then The sub-channels are assigned to multicast group n; The following formula is used to filter the number of unsuitable sub-channels, and the maximum value of PSNR of the previous n multicast groups is solved each time by recursion of the following formula, and finally the spectrum allocation strategy capable of maximizing the aggregate PSNR of the multicast groups is obtained ; 。
2. 2. The online SVC multicast method based on unmanned aerial vehicle relay in NOMA network according to claim 1, wherein in step 1), the clustering algorithm is a mean shift-shift algorithm.
3. 3. The online SVC multicast method based on unmanned aerial vehicle relay in NOMA network according to claim 1, wherein in step 4.2), in the cluster search process of the improved maximum weight cluster algorithm based on branch delimitation, the delimitation program prunes branch vertices that do not meet the upper limit.

Description

Online SVC multicast method based on unmanned aerial vehicle relay in NOMA network Technical Field The invention belongs to the technical field of communication networks, and particularly relates to an online SVC multicast method based on unmanned aerial vehicle relay in a NOMA network. Background With the development of network technology, real-time video services (such as video conferences, live events, etc.) have been integrated into people's lives. Omdia in report Network Traffic Forecast:2019-24, it was anticipated that video would account for over 75% of the total traffic of the wireless network by 2025. The proliferation of real-time video traffic has led to tremendous pressure on network resource allocation. In contrast to unicast, the bandwidth resources used by multicast [1] are not limited by the number of access users. Scalable Video Coding (SVC) [2] [3] encodes video into a base layer and multiple enhancement layers. The device may adjust the number of decoding layers and reconstruct the complete video according to the network environment and decoding capabilities. SVC multicasting has become a potential video multicast quality enhancement scheme due to flexibility and adaptability. Each SVC video layer under Orthogonal Multiple Access (OMA) is transmitted over a different orthogonal channel. Through power domain multiplexing, non-Orthogonal Multiple Access (NOMA) 4 can provide services to multiple terminals on the same channel. When the transmitting end adopts a non-orthogonal mode to transmit different video layers, the receiver can operate Successive Interference Cancellations (SIC) 5 to demodulate the signal according to the strength of the received signal power. In the base station edge region, the link that the user communicates with the base station is typically a non-line-of-sight communication link. Simply relying on base stations for multicasting is not enough to ensure that edge users receive high quality video. With its mobility, the drone can be temporarily deployed to the edge area covered by the base station, its altitude and line of sight link [6] helping to reduce the resource consumption of video transmission. By combining NOMA and SVC, the unmanned aerial vehicle deployed in the base station edge area can share the frequency spectrum with the base station and perform signal superposition on the power domain, so that the fairness and the resource utilization rate of the base station edge multicast service are improved. For improving base station edge video quality with drones, many challenges remain to be addressed: (1) Unmanned aerial vehicle-base station cooperation The video distribution of the existing unmanned aerial vehicle is mostly cache-enabled, namely, offline service [15] [16] is provided for the ground user through carrying a cache, wherein the video cache needs to be updated regularly to improve the cache hit rate [10]. In order to support online services, one method is to establish a link [7] [8] to a base station through mmWave, but the normal communication distance of mmWave is 150 meters [17], which is far smaller than the coverage radius of a macro base station, so that it is difficult for unmanned aerial vehicles at the edge of the base station to establish stable connection. Most existing hierarchical multicast schemes face the ground network. Researchers have explored a NOMA-enabled SVC multicast scheme with base and enhancement layers sent by terrestrial base stations [2]. Under this framework, there has also been proposed a joint power allocation and subpacket scheme that maximizes the aggregate multicast rate while meeting power and rate constraints [9]. (2) Layered video decoding in a multi-drone scenario. The macro base station transmitting base layer and the small base station transmitting enhancement layer are one common layered video multicast strategy [18]. The base layer has higher requirements on transmission rate than the enhancement layer, meaning that the macro base station needs to occupy more bandwidth resources to transmit the base layer. The unmanned plane can transmit the base layer by virtue of the line-of-sight link with less resources. After receiving the base layer and the enhancement layer, the user equipment decodes according to the power strength and the order and eliminates the interference. Unmanned aerial vehicle deployment must consider the receiving sequence of the video layer and the receiving rate of the user side signal, but the existing work does not involve related problems. Researchers have studied full duplex NOMA systems with multiple drones in coordination to improve the aggregate throughput of the system through dynamic user clustering, drone placement and power allocation [12]. There have also been researchers proposed a power distribution and drone trajectory joint optimization strategy to boost the user reception rate [13]. (3) And (3) association selection of the unmanned aerial vehicle-multicast group. Each mul