Resource deployment optimizer for non-geostationary and/or geostationary communications satellites

Abstract

Systems, methods and techniques are presented for discovering optimal solutions to satisfy communication traffic demands to a NGSO and GSO satellite constellations used for telecommunication. When multiple ground demands (mobile and stationary) are present, a satellite constellation requires an assignment of satellite resources to optimally match the ground demands. The systems, methods and techniques presented can utilize an optimization structure to maximize the objective function, using linear programming in combination with simulation and predictive features. The techniques presented determine optimal or quasi-optimal allocation of scarce and highly constrained satellite resources in an efficient manner. These techniques take into account maximizing capacity while protecting other geostationary and non-geostationary networks.

1 . A communication system comprising:
 a constellation of a plurality of non-geostationary and/or geostationary satellites, each of said satellites having assignable communication resources;   a ground system consisting of one or more Earth stations for transmitting to, and receiving signals from, said constellation of satellites;   a plurality of satellite terminals for transmitting to, and receiving signals from, said constellation of satellites; and   a controller operable to dynamically assign satellite and ground system resources in response to demand for communications services required by the plurality of satellite terminals;   wherein the controller is further operable to:
 pre-compile a link budget recipe and compute link budgets for all potential links; and 
 execute an optimization algorithm which uses said pre-computed link budgets to dynamically assign satellite and ground system resources in response to demand for communications services required by the plurality of satellite terminals. 
   
     
     
         2 . The system of  claim 1  wherein each of the plurality of satellite terminals is located in a position selected from the group consisting of:
 in the Earth, 
 in the air, and 
 in orbit. 
 
     
     
         3 . The system of  claim 1  wherein said controller is selected from the group consisting of:
 a centralized controller; 
 a distributed resource controller; and 
 a plurality of controllers. 
 
     
     
         4 . The system of  claim 1  wherein said controller is operable to pre-compile said link budget recipe and pre-compute link budgets using vector-processing. 
     
     
         5 . The system of  claim 4  said vector-processing comprises a low-level parallelization scheme. 
     
     
         6 . The system of  claim 1  wherein the controller is operable to execute an optimization algorithm which assigns satellites by individual grid point or on a cell-by-cell basis, and assigns satellite communication beam resources by individual grid point or on a cell-by-cell basis. 
     
     
         7 . The system of  claim 1  wherein the controller is operable to dynamically allocate satellite resources by performing a Venetian Blind algorithm wherein:
 a demand grid comprising a continuous stream of time-steps, which is divided into two streams of time-steps, the Venetian Blind algorithm alternatingly assigning blocks of said time steps into said first and second stream; 
 each block of time-steps in said first stream of time-steps being optimized in isolation from other blocks of time-steps in said first stream; 
 each block of time-steps in said second stream of time-steps being optimized in isolation from other blocks of time-steps in said second stream, using the optimized blocks of time-steps in said first stream of time-steps as boundary-condition constraints from said optimized blocks in said second stream of time-steps. 
 
     
     
         8 . The system of  claim 1  wherein prior to optimization, the controller is operable to:
 characterize potential satellite to user uplinks and downlinks in terms of spectral efficiency and payload power utilization efficiency, and 
 input said spectral efficiency and payload power utilization efficiency data to the optimization. 
 
     
     
         9 . The system of  claim 1  wherein the controller is operable to dynamically allocate satellite resources by:
 determining a relaxed solution on a point to point basis, using continuous variables; and 
 then solving the original mixed integer problem with a more narrowly defined objective and bounds. 
 
     
     
         10 . The system of  claim 1  wherein
 the controller is operable to dynamically allocate satellite resources by:
 balancing minimum satisfaction and average satisfaction, this weighted objective giving equal weights to maximize the minimum satisfaction and the average satisfaction across all grid points; and 
 performing an optimization calculation using a straight MIP (mixed integer programming) formulation. 
 
 
     
     
         11 . The system of  claim 1  wherein the controller is operable to dynamically allocate satellite resources by:
 balancing average satisfaction and aggregate capacity, this weighted objective giving equal weight to maximize the average satisfaction across all grid points and the aggregate delivered capacity; and 
 performing an optimization calculation using a straight MIP (mixed integer programming) formulation. 
 
     
     
         12 . The system of  claim 1  wherein the controller is operable to dynamically allocate satellite resources by:
 balancing average satisfaction and aggregate revenue, this weighted objective giving equal weight to maximize the average satisfaction across all grid points and the aggregate revenue; and 
 performing an optimization calculation using a straight MIP (mixed integer programming) formulation. 
 
     
     
         13 . The system of  claim 1  wherein the controller is operable to dynamically allocate satellite resources by:
 characterizing the allocation of satellite resources as an optimization problem of integer variables; 
 determining a relaxed solution to the optimization problem by converting integer variables to continuous variables; 
 maximizing the minimum satisfaction, which is the ratio of the given bandwidth to the requested bandwidth; and 
 solving the optimization problem with a better-behaved objective, comprising a minimum bound and a reduced satisfaction solution space. 
 
     
     
         14 . The system of  claim 1  wherein said controller is operable to relax the demand to determine a feasible demand grid which can be met, prior to performing said optimization. 
     
     
         15 . The system of  claim 1  wherein said optimization includes the allocation of throughput on inter-satellite links (ISL). 
     
     
         16 . The system of  claim 1  wherein said controller is operable to dynamically allocate satellite resources including beam configurations, satellite beam pointing and beam hopping schedule, satellite transmit power, channel bandwidth, symbol rates, data rates, and data paths. 
     
     
         17 . The system of  claim 1  wherein said controller is operable to incorporate power flux density masks to protect other networks. 
     
     
         18 . The system of  claim 1  wherein said controller is operable to incorporate power flux density masks, performing power flux density mask calculations as independent tasks, allowing for parallel processing. 
     
     
         19 . The system of  claim 1  wherein said controller is operable to condition demand to a feasible state by responding to customers who demand satellite resources which go beyond the available capacity, by relaxing the demand to define a feasible demand grid that can be met, the feasible demand grid being used as an input to the optimization calculation. 
     
     
         20 . The system of  claim 1  wherein said controller is operable to provide Full link routing and selection optimization, supporting both forward and return links to user terminals routed to a Point of Presence (PoP), by dynamically allocating satellite resources under the additional constraints of:
 aggregate throughput supported by all active links from a given PoP to a given user satellite matching the total throughput delivered to all users assigned to that PoP from this satellite (this applies in both forward downlink and return directions); 
 aggregate throughput of all active links over a given ISL being equal to or less than the ISL throughput capability; and 
 aggregate bandwidth required to support all active links through a landing station/satellite beam not exceeding the total bandwidth assigned through that beam (this applies in both the forward and return directions). 
 
     
     
         21 . The system of  claim 1  wherein said controller is operable to provide Fading analysis and Mitigation as part of the optimization process, by simulating constellation performance using a historical set of globally distributed rain rate data, and optimizing resource allocation using rain fade calculated on the basis of a global rain rate forecasts. 
     
     
         22 . The system of  claim 1  wherein said controller is operable to model beam bandwidth variables using integers to capture the granularity of the allocatable resource. 
     
     
         23 . The system of  claim 1  wherein said controller is operable to manage beam squint by assigning frequencies to terminals based upon their actual location relative to the position of the beam center at the center frequency. 
     
     
         24 . The system of  claim 1  wherein said controller is operable to group terminals into fixed ground cells where cell members are jointly connected to a common satellite. 
     
     
         25 . The system of  claim 1  wherein said controller is operable to optimize resource allocations in a constellation whose satellites can support a constrained set of beam positions, by:
 including an integer constraint of the number of beam positions or targets allowed at a given time; and 
 optionally setting branch priorities based on each beam target's capacity demand to speed up resolution of the mixed-integer problem, as higher-demand beam targets are more likely to have a high impact on node feasibility. 
 
     
     
         26 . The system of  claim 1  wherein said controller is operable to optimize for long-term link availability under fade by:
 characterizing the long-term availability of each instantaneous link; and 
 optimizing the link allocations over a multi-time-step block to maximize the time-averaged long-term availability of the chain of links to any given terminal, in tandem with clear-sky capacity optimization using weighted objectives and/or hierarchical objectives. 
 
     
     
         27 . The system of  claim 1  wherein said controller is operable to:
 optimize resource allocations as a network flow problem where the flow value represents the number of links, approximating satellite constraints by flow-capacity constraints which have unitary coefficients and integer constants; 
 rather than performing a Mixed Integer Linear Programming optimization utilizing binary variables for the assignment of terminals to satellites. 
 
     
     
         28 . The system of  claim 27  wherein said satellite constraints comprise beam bandwidth, frequency reuse constraint, and available radiated RF power. 
     
     
         29 . The system of  claim 1  wherein said controller is operable to perform the optimization using a satellite-view beam layout whose pattern moves along deterministic curves following the same general direction as the apparent movement of a uniform distribution of fixed terminals as seen from the satellite. 
     
     
         30 . The system of  claim 1  wherein said controller is operable to define groups of beams to incorporate frequency re-use constraints, controlling frequency re-use by limiting the aggregate effective bandwidth allocated in any cluster of beams, defined as a group of beams all fully coupled among each other based on a threshold spacing in the satellite field of view. 
     
     
         31 . A method of operation for a satellite system comprising: providing:
 a constellation of a plurality of non-geostationary and/or geostationary satellites, each of said satellites having assignable communication resources;   a ground system consisting of one or more Earth stations for transmitting to, and receiving signals from, said constellation of satellites; and   a plurality of satellite terminals for transmitting to, and receiving signals from, said constellation of satellites; and   dynamically assigning satellite and ground system resources in response to demand for communications services required by the plurality of satellite terminals by:
 pre-compiling a link budget recipe and computing link budgets for all potential links; and 
 executing an optimization algorithm which uses said pre-computed link budgets to dynamically assign satellite and ground system resources in response to demand for communications services required by the plurality of satellite terminals. 
   
     
     
         32 . The method of  claim 31  wherein each of the plurality of satellite terminals is located in a position selected from the group consisting of:
 in the Earth, 
 in the air, and 
 in orbit. 
 
     
     
         33 . The method of  claim 31  wherein said pre-compiling and executing an optimization are performed in a manner selected from the group consisting of:
 in a centralized manner; 
 in a distributed manner; and 
 in a plurality of separate controllers. 
 
     
     
         34 . The method of  claim 31  wherein said step of pre-compiling said link budget recipe and pre-compute link budgets comprises pre-compiling a link budget recipe and compute link budgets for all potential links using vector-processing. 
     
     
         35 . The method of  claim 34  wherein said vector-processing comprises executing a low-level parallelization scheme.