Search

CN-121728513-B - Cache and safe multi-hop partial unloading method in ultra-dense millimeter wave network

CN121728513BCN 121728513 BCN121728513 BCN 121728513BCN-121728513-B

Abstract

The invention discloses a cache and safe multi-hop partial unloading method in an ultra-dense millimeter wave network, which belongs to the field of wireless communication and comprises the steps of obtaining network basic information of an ultra-dense millimeter wave mobile edge computing network supporting non-orthogonal multiple access and orthogonal frequency division multiple access, constructing a network system according to the network basic information and constructing an optimization problem under the constraint of the network system, generating an initial population according to a certain rule, searching the initial population by adopting an improved gradient optimization algorithm to obtain a target population, outputting the position of a historical optimal individual in the target population, and executing safe computing efficiency optimization configuration according to the position of the historical optimal individual. The invention has the advantages of unloading by multiple base stations, considering the cache and millimeter wave factors, meeting the rate, the energy consumption, the time delay and the security hole cost of the mobile terminal and the constraint of the task and the proportion of the transmitting power, and well realizing the aim of minimizing the local energy consumption of all the mobile terminals.

Inventors

  • ZHOU TIANQING
  • ZHOU YUANYANG
  • LI XUAN
  • Nie Xuefang

Assignees

  • 华东交通大学

Dates

Publication Date
20260508
Application Date
20260213

Claims (9)

  1. 1. The method for unloading the buffer memory and the safe multi-hop part in the ultra-dense millimeter wave network is characterized by comprising the following steps: step S1, acquiring network basic information of an ultra-dense millimeter wave mobile edge computing network supporting non-orthogonal multiple access and orthogonal frequency division multiple access and constructing a network system, wherein the network system comprises a communication model, a caching and unloading model, a security model and a computing model, and the optimization problem is constructed under the constraint of the network system, wherein the specific process of constructing the caching and unloading model is as follows: setting a buffer status indicator for each task on the base station, wherein when the value of the indicator is 1, the task is buffered in the base station, and when the value of the indicator is 0, the task is not buffered; When the mobile terminal establishes association with the micro base station, firstly checking whether a task exists in a storage space of the micro base station, if the task is cached in the micro base station, directly processing the whole task in a local base station, and transmitting a processing result back to the mobile terminal through a wireless backhaul link; if the task is cached in the macro base station, the whole task is processed on the macro base station, and a processing result is sent to the micro base station through a wired link and then returned to the mobile terminal through a wireless backhaul link; when the mobile terminal establishes association with the macro base station, firstly checking whether a task exists in a storage space of the macro base station, if the task is cached on the macro base station, directly processing the whole task on the macro base station, and directly transmitting a processing result to the mobile terminal through a wireless backhaul link; Step S2, encoding feasible solutions of the optimization problem; Step S3, generating an initial population according to a certain rule based on a coding result, iterating the initial population by utilizing an improved gradient optimization algorithm, and additionally carrying out diversity-guided mutation operation to update population position information on the basis of gradient search rule operation and local escape operator operation in an original gradient optimization algorithm in the iterative process; And S4, performing task calculation unloading resource allocation according to the position information of the selected optimal individual of the population history.
  2. 2. The method for offloading cache and security multi-hop portions in an ultra-dense millimeter wave network according to claim 1, wherein the ultra-dense millimeter wave mobile edge computing network supporting non-orthogonal multiple access and orthogonal frequency division multiple access comprises macro base stations, a plurality of micro base stations and a plurality of mobile terminals, each micro base station is equipped with a mobile edge computing server, all micro base stations are connected with the nearest macro base station through a wired backhaul link, the number of micro base stations is greater than or equal to the number of mobile terminals, and any mobile terminal is required to complete in turn under the double constraint of security cost and time limit The mobile terminal and a part of micro base stations communicate by millimeter waves, and the rest micro base stations communicate with the mobile terminal by macro waves; 1 macro base station exists in ultra-dense millimeter wave mobile edge computing network supporting non-orthogonal multiple access and orthogonal frequency division multiple access The number of each micro base station and the macro base station is 0, and the index set of each micro base station is And, aggregate Communication between micro base station and mobile terminal by millimeter wave, and the method comprises the following steps of , Representing the use of a collection of micro base stations communicating in millimeter waves between the micro base stations and the mobile terminals, Indicating the number of communication using millimeter waves between the micro base station and the mobile terminal, Representing the set of all base stations including micro base station and macro base station, the index set of the mobile terminal is recorded as Each mobile terminal The item task index set is , Indicating the number of mobile terminals and, Indicating the index number of the mobile terminal, An index number representing a task is provided, Indicating the total number of tasks, mobile terminal Is the first of (2) The task is , Is the task data volume; The number of CPU cycles required for calculating each bit of data, namely the calculation intensity; a desired level of security for the task; Is the maximum safety cost which can be borne; the maximum tolerable processing delay of the task is set; Each micro base station divides the frequency spectrum resource into the frequency spectrum resources based on the principle of equal bandwidth division Orthogonal sub-channels are divided into millimeter wave channels and macro wave channels, and the index set of the orthogonal sub-channels is , , For the orthogonal sub-channel index, Is the sum of millimeter wave channels and macro wave channels; representing the bandwidth of the system and, For orthogonal sub-channel bandwidth, association set The mobile terminals of the middle micro base station are millimeter wave mobile terminals, the mobile terminals associated with the other micro base stations are macro wave mobile terminals, all adopt an orthogonal multiple access mode to share the same micro base station orthogonal sub-channel, and the mobile terminals associated with the macro base station are used for sharing the frequency band resources of the macro base station.
  3. 3. The method for unloading the buffer and the secure multi-hop portion in the ultra-dense millimeter wave network according to claim 2, wherein the specific process of constructing the communication model is as follows: Orthogonal subchannels Mobile terminal associated with macro base station Uplink transmission rate of (a) Expressed as: ; in the formula, Representing the bandwidth of the macro channel; for mobile terminal Other mobile terminals causing worse channel gains of interference; for mobile terminal Other mobile terminals with worse channel gain causing interference Association decisions with a macro base station; representing mobile terminals Is set to the transmission power of (a); is the noise power of the macro wave channel; representing mobile terminals And a channel gain between the macro base station; Orthogonal subchannels Mobile terminal associated with micro base station and communicating using millimeter wave channel Uplink transmission rate of (a) Expressed as: ; in the formula, Representing the bandwidth of the millimeter wave channel; is a mobile terminal Is set to the transmission power of (a); Representation of mobile terminals The transmit power of other mobile terminals that cause worse channel gain of the interference, Is a mobile terminal And micro base station Channel gain for millimeter wave communication between; Representation of mobile terminals Other mobile terminals with worse channel gain causing interference And micro base station Channel gain for millimeter wave communication between; noise power for millimeter wave channels; Indicated in the sub-channel Upper-to-lower mobile terminal And micro base station A mobile terminal index set which generates interference in communication; Representation of mobile terminals Other mobile terminals with worse channel gain causing interference And micro base station The gain of the channel of the communication between them, Is a mobile terminal And micro base station Channel gain for communication between; And Respectively mobile terminals The number of the selected orthogonal sub-channels and micro base stations, And Respectively representing the mobile terminals Other mobile terminals with worse channel gain causing interference Numbering of the selected orthogonal sub-channels and the micro base station; is the cluster to which the micro base station belongs; when meeting the requirements And is also provided with When the mobile terminal is in the same cluster, the mobile terminal can use the orthogonal sub-channels of different micro base stations in different clusters Is subjected to from Is outside the cluster and cross the interference of base station; Orthogonal subchannels Mobile terminal associated with micro base station and communicating using macro wave channel The uplink transmission rate of (a) is expressed as: ; In the formula (I), the total number of the components, Is a mobile terminal And micro base station Channel gain for macro-wave communication between; Representation of mobile terminals Other mobile terminals with worse channel gain causing interference And micro base station Channel gain for macro-wave communication between; when meeting the requirements And is also provided with When the mobile terminal is in the same cluster, the mobile terminal can use the sub-channels of different micro base stations in different clusters Is subjected to from Is outside the cluster and cross the interference of base station; The noise power of the macro wave channel and the noise power of the millimeter wave channel are respectively as follows: ; Wherein, the Representing mobile terminals Association decisions with a macro base station; Mobile terminal And micro base station In sub-channels The channel gain in (a) is given by: ; in the formula, Representing mobile terminals And micro base station In sub-channels Channel gain in (a); representing the path loss at the time of transmission in the channel; Representing the speed of light; Representing carrier frequency in a channel, wherein the carrier frequency is 2GHZ when a macro wave channel and 70GHZ when a millimeter wave channel; the antenna gain of the macro wave channel is 0dBi, when the millimeter wave channel is used as the main lobe gain, the antenna gain value is 18dBi, and when the millimeter wave channel is used as the side lobe gain, the antenna gain value is 2dBi; Representing the small scale component of the channel; Representing the path loss of the channel.
  4. 4. The method for offloading cache and secure multi-hop portions in an ultra-dense millimeter wave network according to claim 1, wherein the optimization problem is expressed as: ; in the formula, Representing the objective function to be optimized, Representing the total energy consumption of the local calculation, For the unified representation of the optimization problem related set, the unified representation is recorded as a variable set; representing mobile terminals Micro base station A set of association decisions between the two, Representing mobile terminals Micro base station Is a correlation decision of (1); is a mobile terminal Is the first of (2) Personal task pair cryptographic algorithm Is a set of security decision indicators of (c), Is a mobile terminal Is the first of (2) Personal task pair cryptographic algorithm Is used as an indication of the security decision of (a), For the set of indices of the cryptographic algorithm, An index number representing the cryptographic algorithm, Representing the total number of cryptographic algorithms; is a mobile terminal In the orthogonal sub-channels The association on the set of indications, Is a mobile terminal In the orthogonal sub-channels An associated indication thereon; A frequency band division factor; is a mobile terminal A set of transmit powers is provided which, Is a mobile terminal A transmit power; is a mobile terminal Is provided with a local set of computing capabilities, Is a mobile terminal Is provided; is a mobile terminal Is the first of (2) Offloading individual tasks to a micro base station Is used for the data volume collection of (a), Is a mobile terminal Is the first of (2) Offloading individual tasks to a micro base station Is a data amount of (a); is a mobile terminal Is the first of (2) Personal task via micro base station Decryption And then, unloading the data volume set to the macro base station again, Is a mobile terminal Is the first of (2) Personal task via micro base station Decryption Then, unloading the data quantity to the macro base station again; for the task At a base station A set of cache state indicators on the cache memory, For the task At a base station A cache status indicator thereon; meaning of constraint: is time delay constraint and represents the completion time of all tasks of the mobile terminal Must not exceed its maximum tolerable delay ; Is a security constraint representing the total security breach costs of the mobile terminal Must not exceed its preset upper limit ; And (3) with Together forming an association constraint, which means that each mobile terminal can only associate with one base station; and (3) with The encryption algorithm constraint is formed together, which means that each task can only select one encryption algorithm; and (3) with Together forming a subchannel constraint, meaning that each subchannel is assigned to at most one mobile terminal; and (3) with Forming a buffer quantity constraint together, wherein the buffer quantity constraint represents that each task is buffered in one base station at most; is a buffer total constraint, which indicates that the buffer task total amount of the base station does not exceed the maximum buffer capacity ; Is a computational capability constraint representing the local computational capability of the mobile terminal The upper limit is ; Is a band division factor constraint, representing a band division factor The value range is (0, 1); Is a transmit power constraint representing the transmit power of the mobile terminal The upper limit is ; Is an offload data volume constraint, representing that the multi-step offload data volume satisfies , Is a mobile terminal Is the first of (2) Total offloaded data volume for each task.
  5. 5. The method for buffering and secure multi-hop partial offload in an ultra-dense millimeter wave network of claim 4, wherein step S2 is expressed as: Will optimize the problem The individual possible solutions are encoded as individual vectors, and the index set is For any vector individual The whole optimization problem is related to be gathered 、 、 、 、 、 、 、 And One-to-one mapping into multiple component sets: as vector individuals Mobile terminal The set of base station indices selected is selected, As vector individuals Mobile terminal A selected base station index; as vector individuals Mobile terminal The selected cipher algorithm numbers the set, As vector individuals Mobile terminal The number of the selected cipher algorithm is selected, A mobile terminal set containing different tasks; as vector individuals Mobile terminal The set of assigned sub-channel indices, As vector individuals Mobile terminal An assigned subchannel index; as vector individuals Frequency band division factors of (2); as vector individuals Mobile terminal Is provided with a set of computing capabilities of (a), As vector individuals Mobile terminal Is used for the calculation of the calculation capacity of (a); as vector individuals Mobile terminal The selected set of cached base station indices, As vector individuals Mobile terminal A selected cache base station index; as vector individuals Mobile terminal Is set of transmit powers of (a), As vector individuals Mobile terminal Is set to the transmission power of (a); as vector individuals Mobile terminal The set of data volumes offloaded to the associated base station, As vector individuals Mobile terminal The amount of data offloaded to the associated base station; as vector individuals Mobile terminal The associated set of data volumes forwarded by the micro base station to the macro base station, As vector individuals Mobile terminal The associated micro base station forwards the amount of data to the macro base station.
  6. 6. The method for offloading a cache and a secure multi-hop portion of an ultra-dense millimeter wave network according to claim 5, wherein the specific operation in step S3 is as follows: Step S3.1, setting the maximum iteration times , Starting from 1 for the current iteration number; S3.2, initializing population individuals, and generating an initial population according to the following rules; ; in the formula, Representing vector individuals Other mobile terminals in (a) An initial solution of the selected base station index, As vector individuals Other mobile terminals in (a) An initial solution of the assigned subchannel index, As vector individuals Is a solution to the initial frequency division factor of (c), As vector individuals Other mobile terminals in (a) Is an initial solution to the computing power of (a), As vector individuals Other mobile terminals in (a) Is a solution to the initial solution of the transmit power of (a), As vector individuals Mobile terminal related to task The initial solution of the number of the selected cipher algorithm, As vector individuals Mobile terminal related to task An initial solution of the selected cache base station index, As vector individuals Mobile terminal related to task The amount of data offloaded to the associated base station, As vector individuals Mobile terminal related to task The amount of data that the associated micro base station forwards to the macro base station, Representation from discrete sets Uniformly and randomly selecting an element; Is shown in the interval Generating a real number at random; representing individual mapping of linear index vectors to Row subscript for matrix And column subscripts ; Representing performing a random retention operation in a capacity space on the input variable; Step S3.3, setting the current position of the individual in the t-th iteration as Performing gradient search rule operation on the population to obtain the updated position of the individual through the gradient search rule operation The process is as follows: ; in the formula, , To contain 4 component sets Is a first set of encoding variables of (c), To contain 4 component sets Is provided with a second set of encoding variables, For the current individual position of the t-th iteration of the first set of encoding variables, For the current individual position of the t-th iteration of the second set of encoding variables, , Operating the updated positions of the individuals through gradient search rules for the first coding variable set and the second coding variable set respectively; , the historic worst individual and historic best individual positions after the t-th iteration of the first set of coding variables are respectively, , Respectively obtaining the historic worst individual and historic best individual positions after the t-th iteration of the second coding variable set; representing a rounding function; representing a standard normal random number; Representing a random number between (0, 1); Is a small constant; is element by element, In order to divide the elements by elements, , A first adaptive iteration coefficient and a second adaptive iteration coefficient, respectively, which are equal in value, The third gradient search direction and the second gradient search direction are respectively; and then introducing Newton method ideas to update the population positions, wherein the position updating rule is as follows: ; Wherein: , To introduce updated positions of individuals in newton's method by gradient search rule operations, , Introducing updated positions of the Newton's method individuals to the first encoding variable set and the second encoding variable set through gradient search rule operation, The first and third gradient coefficients respectively introducing newton's method, Introducing second and fourth gradient coefficients of Newton's method respectively; individual current position based on the t-th iteration Two new forms updated positions And (3) with Next generation vector individual Is determined in the following manner: ; Wherein, the , A first coding variable set and a second coding variable set after the t+1st iteration respectively, Representing randomly generating a real number within the interval (0, 1); step S3.4, carrying out compound operation on the population, enabling the whole population to sequentially execute gradient search rule and local escape operator operation, carrying out boundary processing and greedy selection after each operation, and updating position information with worst history and best history, wherein the specific process is as follows: With probability For individual position after the t+1st iteration The following operations are performed: ; ; Wherein, the , Vector individuals after the t-th iteration of the first coding variable set respectively Is provided in the position of (a), Individual random vectors after the t-th iteration for the first set of encoding variables The location of the update is then updated and, Vector individuals after the t-th iteration of the second coding variable set respectively Is provided in the position of (a), Individual random vectors after the t-th iteration for the second set of encoding variables The location of the update is then updated and, Is interval of Random numbers uniformly distributed in the inner part; random numbers are distributed for standard normal states; Is that A random number within; And S3.5, carrying out diversity-guided mutation operation on the population, wherein the specific process is as follows: Diversity measurement of multi-dimensional numerical problems The definition is as follows: ; in the formula, Representing the optimization problem related sets respectively 、 、 、 、 、 Diagonal length of feasible region; as vector individuals Mobile terminal The average value of the base station index selected, As vector individuals Mobile terminal The selected cipher algorithm numbers the average value, As vector individuals Mobile terminal The average value of the assigned sub-channel index, As vector individuals Mobile terminal Is a mean value of the computing power of (c), As vector individuals Mobile terminal The average value of the selected cache base station indexes, As vector individuals Mobile terminal Is a mean value of the transmit power of (a); performing mutation operation according to probability, namely ; In the formula, All are constant coefficients greater than 0 and less than 1, representing a diversity threshold, For the first, second and third probability of variation, Is the probability of performing a mutation operation, including 3, When the variation probability of the diversity threshold meets the requirement, performing a variation operation on individuals in the population again; Step S3.6, judging whether the current iteration number reaches the maximum iteration number, if not, executing the steps S3.3, S3.4 and S3.5 in sequence, and adding one to the iteration number after the execution is finished; and S3.7, calculating the fitness value of all the individuals in the current population by using the fitness function, selecting the individual with the highest fitness value as the current optimal individual, comparing the current optimal individual with the initialized optimal individual, taking the individual with the large fitness value as the historical optimal individual, and outputting corresponding position information.
  7. 7. The method for cache and secure multi-hop partial offload in an ultra-dense millimeter wave network of claim 6, wherein vector individuals The fitness function of (2) is defined as: ; in the formula, As vector individuals Is a set of encoding variables of (a); Is a fitness function; and (3) with Respectively mobile terminals Penalty coefficients corresponding to the latency constraints and the security constraints of the system.
  8. 8. A non-volatile computer storage medium having stored thereon computer executable instructions for performing the cache and secure multi-hop partial offload method in an ultra-dense millimeter wave network as claimed in any one of claims 1 to 7.
  9. 9. An electronic device comprising at least one processor and a memory communicatively coupled to the at least one processor, wherein the memory stores instructions executable by the at least one processor, wherein the instructions are executable by the at least one processor to cause the at least one processor to perform the cache and secure multi-hop portion offload method in an ultra-dense millimeter wave network of any of claims 1-8.

Description

Cache and safe multi-hop partial unloading method in ultra-dense millimeter wave network Technical Field The invention belongs to the field of wireless communication, and particularly relates to a cache and safe multi-hop part unloading method in an ultra-dense millimeter wave network. Background Iteration of communication technology and popularization of terminals promote a large number of computationally intensive and time-delay harsh applications such as face recognition, natural language processing, interactive games and the like, but highlight energy consumption bottlenecks, namely limitation to factors such as chip computing power and the like, and local equipment often runs. The moving edge calculation (moving edge calculation, mobile Edge Computing) is straightforward to offload, while millimeter waves (mmWave, MILLI METER WAVE) are seen as an effective means of further "latency and acceleration rate". However, millimeter wave natural coverage is short, blind can be supplemented only by ultra-dense base stations, so that the energy consumption of the whole network is increased suddenly, and meanwhile, the malicious invasion risk is amplified by an edge side open interface. In addition, in order to reduce the empty and return costs caused by repeated unloading, a 'task level buffer memory' is introduced at the edge side, which not only buffers static contents, but also pre-stores hot reusable tasks and intermediate data thereof in the edge cloud. Once the mobile terminal requests for hit, the edge node directly executes locally and returns the result without uploading the original task again, thereby further reducing time delay, saving bandwidth, expanding the system and reducing the energy consumption of the local terminal. Therefore, under the framework of ultra-dense millimeter wave and Non-orthogonal multiple access (NOMA, non-Orthgonal Multiple Access) +orthogonal frequency division multiple access (OFDMA, orthogonal Frequency Division Multiple Access), a green, safe, multi-step and buffer unloading strategy is provided, wherein the combined optimization of task unloading and buffer is realized, so that the service quality is ensured, and the energy consumption is minimized. In order to solve the above problems, under multiple constraint conditions such as time delay and security cost, how to reduce local energy consumption and signal interference of the mobile terminal is sought, so as to improve user experience. The traditional gradient optimization algorithm has the problems of limited global exploration capacity, easy sinking into a local optimal solution and the like, and has a certain limitation in practical application, so that the algorithm is required to be improved so as to improve the performance and applicability of the algorithm. Disclosure of Invention Aiming at the defects of the prior art, the invention provides a cache and safe multi-hop part unloading method in an ultra-dense millimeter wave network, which aims to solve the problems discussed in the background art. In order to achieve the above purpose, the present invention provides the following technical solutions. A method for unloading a cache and a safe multi-hop part in an ultra-dense millimeter wave network comprises the following steps: Step S1, acquiring network basic information of an ultra-dense millimeter wave mobile edge computing network supporting non-orthogonal multiple access and orthogonal frequency division multiple access and constructing a network system, wherein the network system comprises a communication model, a caching and unloading model, a security model and a computing model; Step S2, encoding feasible solutions of the optimization problem; Step S3, generating an initial population according to a certain rule based on a coding result, iterating the initial population by utilizing an improved gradient optimization algorithm, and additionally carrying out diversity-guided mutation operation to update population position information on the basis of gradient search rule operation and local escape operator operation in an original gradient optimization algorithm in the iterative process; And S4, performing task calculation unloading resource allocation according to the position information of the selected optimal individual of the population history. Further, the ultra-dense millimeter wave mobile edge computing network supporting non-orthogonal multiple access and orthogonal frequency division multiple access comprises a macro base station, a plurality of micro base stations and a plurality of mobile terminals, wherein each micro base station is provided with a mobile edge computing server, all micro base stations are connected with the nearest macro base station through a wired backhaul link, the number of the micro base stations is greater than or equal to that of the mobile terminals, and any mobile terminal is required to finish the steps in sequence under the double constraint of safety cost and t