Mobile-assisted edge computing framework for emerging IoT applications

{
  "publication_number": "US11412061B1",
  "country": "US",
  "kind": "B1",
  "title": "Mobile-assisted edge computing framework for emerging IoT applications",
  "abstract": "A Mobile-Assisted edge computing framework including: one or more requests to be processed; an operator configured to assign the one or more requests to a static edge node; and a cloud configured to cache and pre-fetch services for the one or more requests; wherein the one or more requests include handled requests and unhandled requests, the static edge node is configured to process the handled requests, and one or more mobile edge nodes are configured to process the unhandled requests; and wherein the static edge node is configured to auction the unhandled requests to the one or more mobile edge nodes, the auction including: generating a candidate set of the one or more mobile edge nodes; assigning an unhandled request to a target mobile edge node of the candidate set; and processing the unhandled request via the target mobile edge node.",
  "claims": [
    "1. A Mobile-Assisted edge computing framework, comprising: one or more requests to be processed; a static edge node comprising an access point (AP) and an edge server; one or more mobile edge nodes; an operator configured to assign the one or more requests to the static edge node; and a cloud configured to cache and pre-fetch services for the one or more requests; wherein the one or more requests comprise handled requests and unhandled requests, the static edge node is configured to process the handled requests, the unhandled requests comprise requests unable to be processed by the static edge node, the one or more mobile edge nodes are configured to process the unhandled requests, and the static edge node is configured to auction the unhandled requests to the one or more mobile edge nodes, the auction comprising: generating a candidate set of the one or more mobile edge nodes; assigning an unhandled request to a target mobile edge node of the candidate set; and processing the unhandled request via the target mobile edge node; wherein the generating comprises: selecting a subset of requests to be processed at a time slot from the one or more requests; receiving bids for the subset from the one or more mobile edge nodes; comparing a remote cloud task completion time and a mobile edge node task completion time; and generating the candidate set based on the bids and the comparing.",
    "2. The framework of claim 1, wherein the bids are calculated by the one or more mobile edge nodes according to the mobile edge node task completion time, and the mobile edge node task completion time is calculated according to: T i t ⁢ m t = I i t / D i t ⁢ n i t + I i t / D n i t ⁢ m t + K i t / c m t + O i t / D m t ⁢ n i t + O i t / D n i t ⁢ i t wherein T i t m t is the mobile edge node task completion time, i t denotes the unhandled request of a corresponding end user, m t denotes a mobile edge node, I i t is an input data size of the unhandled request, K i t is an amount of required computational resources required for the unhandled request, O i t is an output data size of the unhandled request, c m t is an available processing capacity of the mobile edge node, D i t ⁢ n i t is a user-to-edge transmission rate, and D n i t ⁢ m t is an edge-to-mobile edge node transmission rate.",
    "3. The framework of claim 2, wherein I i t / D i t ⁢ n i t + I i t / D n i t ⁢ m t is a task uploading time, K i t /c m t is a task processing time, and O i t / D m t ⁢ n i t + O i t / D n i t ⁢ i t is a task downloading time.",
    "4. The framework of claim 3, wherein the assigning comprises: constructing a weighted bipartite graph G comprising subgraphs G n according to the static edge node, where G={G 1, G 2,..., G N } and N is the number of subgraphs; and solving a Maximum Matching (MM) problem using the weighted bipartite graph.",
    "5. The framework of claim 4, wherein the weighted bipartite graph is constructed according to a set of the one or more user requests, the candidate set, and a profit of the static edge node.",
    "6. The framework of claim 5, wherein the profit of the static edge node is calculated according to: profit i t ⁢ m t = { v i t - p i t, if ⁢ ⁢ T i t ⁢ m t ≤ T i t DDL, v i t - p i t - F p i t ⁡ (T i t ⁢ m t - T i t DDL), if ⁢ ⁢ T i t ⁢ m t > T i t DDL. where profit i t m t is the profit of the static edge node, v i t is a valuation of the unhandled request for the static edge node, p i t is a price of the unhandled request, T i t DDL is a deadline of task completion time, and F p i t is a penalty function.",
    "7. The framework of claim 5, where the MM problem is solved using a Kuhn-Munkres (KM) algorithm.",
    "8. The framework of claim 7, wherein after the processing, the target mobile edge node is configured to submit a Proof of Participation (PoP) to the static edge node, wherein the PoP comprises an equipment identification number used to identify and track unique mobile edge nodes.",
    "9. The framework of claim 8, wherein when the target mobile edge node submits the PoP but does not complete the unhandled request, a reputation of the target mobile edge node is damaged.",
    "10. The framework of claim 8, wherein when the target mobile edge node does not complete the unhandled request within a guaranteed time, the target mobile edge node is not compensated.",
    "11. The framework of claim 8, wherein Proof of Assignment (PoA) is sent to the target mobile edge node before the processing and the target edge node is compensated for the processing by the static edge node based on the PoA.",
    "12. The framework of claim 1, wherein each mobile edge node of the one or more mobile edge nodes is a smart car, unmanned aerial vehicle (UAV) or robot.",
    "13. A method for processing one or more requests using a Mobile-Assisted computing framework comprising the following steps: generating a candidate set of one or more mobile edge nodes; assigning an unhandled request to a target mobile edge node of the candidate set; and processing the unhandled request via the target mobile edge node; wherein the Mobile-Assisted computing framework comprises: one or more requests to be processed; a static edge node comprising an access point (AP) and an edge server; one or more mobile edge nodes; an operator configured to assign the one or more requests to the static edge node; and a cloud configured to cache and pre-fetch services for the one or more requests; wherein the one or more requests comprise handled requests and unhandled requests, the static edge node is configured to process the handled requests, the unhandled requests comprise requests unable to be processed by the static edge node, and the one or more mobile edge nodes are configured to process the unhandled requests; and the static edge node is configured to auction the unhandled requests to the one or more mobile edge nodes; and wherein the generating comprises: selecting a subset of requests to be processed at a time slot from the one or more requests; receiving bids for the subset from the one or more mobile edge nodes; comparing a remote cloud task completion time and a mobile edge node task completion time; and generating the candidate set based on the bids and the comparing.",
    "14. The method of claim 13, wherein the bids are calculated by the one or more mobile edge nodes according to the mobile edge node task completion time, and the mobile edge node task completion time is calculated according to: T i t ⁢ m t = I i t / D i t ⁢ n i t + I i t / D n i t ⁢ m t + K i t / c m t + O i t / D m t ⁢ n i t + O i t / D n i t ⁢ i t wherein T i t m t is the mobile edge node task completion time, i t denotes the unhandled request of a corresponding end user, m t denotes a mobile edge node, I i t is an input data size of the unhandled request, K i t is an amount of required computational resources required for the unhandled request, O i t is an output data size of the unhandled request, c m t is an available processing capacity of the mobile edge node, D i t ⁢ n i t is a user-to-edge transmission rate, and D n i t ⁢ m t is an edge-to-mobile edge node transmission rate; and I i t / D i t ⁢ n i t + I i t / D n i t ⁢ m t is a task uploading time, K i t /c m t is a task processing time, and O i t / D m t ⁢ n i t + O i t / D n i t ⁢ i t is a task downloading time.",
    "15. The method of claim 14, wherein the assigning comprises: constructing a weighted bipartite graph G comprising subgraphs G n according to the static edge node, where G={G 1, G 2,..., G N } and N is the number of subgraphs; and solving a Maximum Matching (MM) problem using the weighted bipartite graph; wherein the weighted bipartite graph is constructed according to a set of the one or more user requests, the candidate set, and a profit of the static edge node; and the MM problem is solved using a Kuhn-Munkres (KM) algorithm.",
    "16. The method of claim 15, wherein the profit of the static edge node is calculated according to: profit i t ⁢ m t = { v i t - p i t, if ⁢ ⁢ T i t ⁢ m t ≤ T i t DDL, v i t - p i t - F p i t ⁡ (T i t ⁢ m t - T i t DDL), if ⁢ ⁢ T i t ⁢ m t > T i t DDL. where profit i t m t is the profit of the static edge node, v i t is a valuation of the unhandled request for the static edge node, p i t is a price of the unhandled request, T i t DDL is a deadline of task completion time, and F p i t is a penalty function.",
    "17. The method of claim 16, wherein after the processing, the target mobile edge node is configured to submit a Proof of Participation (PoP) to the static edge node, wherein the PoP comprises an equipment identification number used to identify and track unique mobile edge nodes.",
    "18. The method of claim 17, wherein when the target mobile edge node submits the PoP but does not complete the unhandled request, a reputation of the target mobile edge node is damaged.",
    "19. The method of claim 17, wherein Proof of Assignment (PoA) is sent to the target mobile edge node before the processing and the target edge node is compensated for the processing by the static edge node based on the PoA.",
    "20. The method of claim 1, wherein each mobile edge node of the one or more mobile edge nodes is a smart car, unmanned aerial vehicle (UAV) or robot."
  ],
  "description_excerpt": "The present disclosure relates to the field of edge computing, and more particularly relates to a Mobile-Assisted edge computing framework.\n\nSome references, which may include patents, patent applications and various publications, are cited and discussed in the description of this disclosure. The citation and/or discussion of such references is provided merely to clarify the description of the present disclosure and is not an admission that any such reference is “prior art” to the disclosure described herein. All references cited and discussed in this specification are incorporated herein by reference in their entireties and to the same extent as if each reference is individually incorporated by reference. In terms of notation, hereinafter, [n] represents the nth reference cited in the reference list. For example, [1] represents the first reference cited in the reference list, namely, GMSA, “GSMA intelligence,” https://www.gsmaintelligence.com/, Accessed On: February 2019.\n\nAs Internet of Things (IoT) devices become increasingly prevalent, new IoT applications such as mobile gaming, video processing, and image recognition have emerged and are frequently used [1]. The IoT applications are often delay sensitive, resource hungry and computation intensive. However, the IoT devices tend to be designed to be more flexible and easier to transport, which may limit the processing power of the devices. In response, edge computing (EC) may be used as a paradigm that caches services and provides cloud-like resources via static edge nodes (i.e., edge clouds, cloudlets, follow me clouds, fogs, etc.) at the network edge [2].",
  "cpc": [
    "H04W 4/70",
    "H04L 67/12",
    "H04L 67/2847",
    "H04L 67/32",
    "H04L 67/562",
    "H04L 67/5681",
    "H04L 67/60",
    "H04W 4/50"
  ],
  "ipc": [
    "H04L 67/12",
    "H04L 67/5681",
    "H04L 67/60",
    "H04W 4/50"
  ],
  "assignees": [
    "National University of Defense Technology"
  ],
  "inventors": [
    "Deke Guo",
    "Long Zheng",
    "Lailong Luo",
    "Xinyi Li",
    "Siyuan GU"
  ],
  "filing_date": "2021-12-03",
  "publication_date": "2022-08-09",
  "grant_date": "2022-08-09",
  "priority_date": "2021-12-03",
  "application_number": "US-202117542227-A",
  "family_id": "82706065",
  "cited_by_count": 10,
  "citations": [
    "US20170287062A1",
    "US20180139726A1",
    "US20200195497A1",
    "US20210021431A1",
    "US20210119962A1"
  ]
}