BR112016021568B1 - METHOD FOR EFFECTIVE AUTOMATIC CHANGING FOR LOW EARTH ORBIT (LEO) SATELLITE SYSTEMS AND ACCESS NODE IN A MOBILE WIRELESS COMMUNICATIONS NETWORK - Google Patents

Abstract

APPARATUS AND METHOD FOR EFFECTIVE AUTOMATIC CHANGING FOR LOW EARTH ORBITAL (LEO) SATELLITE SYSTEMS. These are approaches to continuous, dynamic, and effective auto-changing processes, which encompasses the selection of an optimal trajectory (consisting of a satellite, a carrier frequency set, and a satellite beam) over which a mobile user terminal (UT) communicates with the radio access network in a mobile satellite communications system. A set of path factors is determined with respect to each of a plurality of communications paths for the UT. A path selection metric (PSM) for each communications path is determined, where the PSM for each communications path is determined through a weighted calculation based on the respective set of path factors for the communications path. A decision is made to determine whether to perform an automatic change of the UT from a first among the communications paths to a second among the communications paths, the determination being based on an assessment carried out based at least in part on the PSM.

Description

BACKGROUND

[001] In a cellular wireless communications system, an automatic change or handover of a user terminal (UT) or handset is a process of transferring an ongoing call or data session from a channel connected to the core network to another channel (e.g., from a cell serviced by a cell tower to another cell served by another cell tower). In a mobile satellite communications system, automatic switching is the reassignment or reconfiguration of radio links between the UT and a satellite, a satellite beam and triplet or carrier frequency set. The criteria for performing the automatic shift may vary from system to system, but in general, automatic shifts are triggered to select trajectories with better channel conditions.

[002] Among the different types of mobile satellite communications systems, satellites orbiting at different levels above the Earth can be employed, such as low earth orbit (LEO) satellites, medium earth orbit (MEO) satellites, and geosynchronous earth orbit (GEO) satellites. A GEO satellite orbits the Earth by flying in a geosynchronous Earth orbit, with an orbital period equal to the Earth's rotation period. A geosynchronous orbit is at an altitude of about 35,786 km (22,236 miles) above the Earth and is typically directly above the Equator. A LEO satellite typically flies in an orbit around the Earth at an altitude between 160 kilometers or 99 between 160 kilometers or 99 miles (with an orbital period of about 88 minutes) and 2,000 kilometers or 1,200 miles (with an orbital period of about 127 minutes). Additionally, in a GEO satellite system, the beam projection area on Earth is generally fixed based on the satellite antenna configuration. Unlike a GEO satellite, a LEO satellite constellation provides a unique mode of operation and associated challenges, where the projection area of each satellite beam on Earth is constantly in motion. Consequently, mobile satellite systems employing LEO satellites require frequent automatic shifts due to beam movement across the Earth (even if there is no terminal movement and the channel condition is constant). In addition, the line of sight between the terminal and the satellite may change due to the movement of the satellite.

[003] Therefore, approaches are needed for continuous, dynamic and effective automatic switching processes, which encompass the selection of an optimal trajectory (consisting of a satellite, a carrier frequency set and satellite beam) over which a user terminal (UT) communicates with the radio access network in a mobile satellite communications system.

SOME EXEMPLIFIER MODALITIES

[004] The present invention advantageously addresses the previously mentioned needs and requirements, as well as others, by providing approaches for continuous, dynamic and effective automatic switching processes, which encompass the selection of an ideal trajectory (consisting of a satellite, a set of carrier frequency and satellite beam) over which a user terminal (UT) communicates with the radio access network in a mobile satellite communications system. The satellite, satellite beam, and carrier frequency set for the forward and reverse link directions do not need to be the same, and no symmetry is assumed in forward and reverse link direction beam patterns. The approaches provided employ a set of metrics to select an optimal trajectory, balance multiple parameters such as signal-to-noise ratio, expected duration of beam coverage, and other parameters, described in further detail below. The approach additionally provides continuous automatic switching from current trajectories to alternative trajectories considered more ideal, in which no loss of transmission data is experienced, except when an automatic change involves an automatic change of carrier frequency.

[005] In accordance with exemplary embodiments of the present invention, a method is provided for evaluating the probability of performing an automatic switch of a UT from a first communications path to a second communications path of a mobile satellite communications system. A set of path factors is determined with respect to each of a plurality of communications paths for the UT. A path selection metric (PSM) for each communications path is determined, where the PSM for each communications path is determined through a weighted calculation based on the respective set of path factors for the communications path. A decision is made to determine whether to perform an automatic change of the UT from a first among the communications paths to a second among the communications paths, the determination being based on an assessment carried out based at least in part on the PSM. In accordance with additional exemplary embodiments, a combined path metric (CPM) is determined for each of a set of candidate communications paths for which the UT may undergo rapid change, wherein the CPM for each set of candidate communications paths is determined based at least in part on the PSM for each communications path in the set. Additionally, each CPM is computed to serve as an indicator reflecting a quality of path diversity over the set of communications paths.

[006] As an example, the set of path factors in relation to each of the communications paths comprises two or more aspects of the communications path, where the aspects comprise one or more indicators that each reflect a quality of the communications path, an expected service life of the communications path, a trend of automatic change and a capacity factor. By way of further example, one or more indicators that each reflect a quality of the communications path comprise a relative pilot strength for the path, the expected lifetime of the communications path comprises an expected lifetime of a subbeam of the communications path, the autoshift bias refers to the timing of the autochange, and the capacity factor reflects a potential transmission capacity of the communications path. By way of further example, the weighted calculation for determining the PSM for each of the communications paths is computed based on a respective different weighting factor applied to each path factor for the communications path, and each different weighting is set to adjust a level of effect of the respective path factor on the PSM to address trade-offs between system operating characteristics.

BRIEF DESCRIPTION OF THE DRAWINGS

[007] The exemplary embodiments of the present invention are illustrated by way of example and not for limiting purposes in the Figures of the attached drawings and in which the similar reference numerals refer to the similar elements, and in which: Figure 1 illustrates a diagram of a communication system capable of providing approaches for dynamic and effective automatic terminal switching in a wireless communication system, in accordance with the exemplary embodiments of the present invention; Figure 2 illustrates a diagram representing sample beam patterns of two respective satellites moving in different directions, with a user terminal (UT) situated in an overlapping section of the two beam patterns, in accordance with exemplary embodiments of the present invention; Figure 3 illustrates a signal timing diagram representing a process for physical channel reconfiguration of frequencies or forwarding paths, for an automatic user terminal (UT) change situation in a CDMA wireless communication system, in accordance with exemplary embodiments of the present invention; Figure 4 illustrates a signal timing diagram that represents a process for physical channel reconfiguration of frequencies or return paths, for an automatic user terminal (UT) change situation in a CDMA wireless communication system, in accordance with exemplary embodiments of the present invention; Figure 5 illustrates a signal timing diagram that represents a process for physical channel reconfiguration of frequencies or return and forwarding paths, for an automatic user terminal (UT) change situation in a TDMA wireless communication system, in accordance with exemplary embodiments of the present invention; Figure 6 illustrates an uplink allocation timing diagram for transmissions relative to nominal positions, for an automatic user terminal (UT) switch situation in a TDMA wireless communications system, in accordance with exemplary embodiments of the present invention; Figure 7 illustrates a diagram representing transmission propagation delays for an automatic user terminal (UT) switching situation in a TDMA wireless communication system, in accordance with exemplary embodiments of the present invention; Figure 8 illustrates a signal timing diagram for an intersatellite auto switch process for a user terminal (UT) auto switch situation in a TDMA wireless communications system, in accordance with exemplary embodiments of the present invention; Figure 9 illustrates a signal timing diagram representing an inter-RNC auto-switching process, for a user terminal (UT) auto-switching situation in a wireless communication system, in accordance with exemplary embodiments of the present invention; and Figure 10 illustrates a signal timing diagram representing an intersatellite and inter-RNC automatic switching process, for an automatic user terminal (UT) switching situation in a wireless communications system, in accordance with exemplary embodiments of the present invention.

DETAILED DESCRIPTION

[008] Approaches to continuous, dynamic and effective automatic switching processes are described, which cover the selection of an ideal trajectory (consisting of a satellite, a set of carrier frequency and satellite beam) over which a user terminal (UT) communicates with the radio access network in a mobile satellite communications system. In the description that follows, for purposes of explanation, several specific details are presented in order to provide a complete understanding of the exemplary embodiments of the invention. It will be apparent to one skilled in the art, however, that additional modalities can be practiced without these specific details or with an equivalent disposition. In other examples, well-known structures and devices are shown in block diagram form in order to avoid unnecessary blurring of embodiments of the invention. Additionally, as will be noted, a module or component (as mentioned herein) may be composed of software component(s), which are stored in memory or other computer-readable storage medium, and executed by one or more processors or CPUs of the respective devices. As will also be noted, however, a module may alternatively be composed of hardware component(s) or firmware component(s), or a combination of hardware, firmware, and/or software components. Additionally, with respect to the various exemplary embodiments described herein, while certain functions are described as being performed by certain components or modules (or combinations thereof), such descriptions are provided as examples and are therefore not intended to be limiting. Consequently, any such functions can be envisioned to be performed by other components or modules (or combinations thereof), without deviating from the general spirit and scope of the present invention.

[009] Additionally, various forms of computer-readable media may be involved in providing instructions to a processor for execution. For example, instructions for carrying out at least part of the present invention may initially be supported on a magnetic disk of a remote computer. In such a scenario, the remote computer loads the instructions into main memory and sends the instructions over a telephone line using a modem. A modem on a local computer system receives the data on the telephone line and uses an infrared transmitter to convert the data to an infrared signal and transmit the infrared signal to a portable computing device such as a personal digital assistant (PDA) and a laptop computer. An infrared detector on the handheld computing device receives the information and instructions supported by the infrared signal and places the data onto a bus. The bus carries data to main memory, from which a processor retrieves and executes instructions. Instructions received by main memory can optionally be stored on the storage device before or after execution by the processor.

[010] Figure 1 illustrates a diagram of a communications system capable of providing approaches for dynamic and effective automatic terminal switching in a wireless communications system, in accordance with exemplary embodiments of the present invention. The terrestrial radio access network (RAN) 113 generally handles all automatic switches. The RAN is the element in a mobile communications network that deploys radio access technology from the underlying physical connection methods to a radio-based communications network. The RAN generally provides the connection between the user terminal (UT) 111/114 (e.g., a mobile telephone) and the core network (CN) 101. The CN comprises the central part of a telecommunications network, which generally provides services (for example, call forwarding via a public service telephone network (PSTN) 105 to end users using the telecommunications network via an access network 103). The exemplary embodiments additionally provide approaches for approximating auto-change factors to evaluate available trajectories and make auto-change decisions when the radio access network (RAN) cannot acquire complete information needed to evaluate all factors. Furthermore, in accordance with the exemplary embodiments, automatic forward and reverse link changes are evaluated independently, and no symmetry is assumed between the forward and reverse path beam patterns and the evaluation of path metrics.

[011] With further reference to Figure 1, the system 100 enables automatic switching procedures between the ground base station and the SBSS 107 to be performed through a core network 101 with standard procedures defined in ground systems. In the example system 100, the UT 111/114 has the capability to communicate over a satellite link (via satellite 109 and satellite base station subsystem (SBSS) or satellite gateway (GW) 107) to wireless core network (CN) 101, or communicate over a terrestrial wireless cell station or tower 110 with a terrestrial radio access network (RAN) 113 to wireless core network (CN) 101. The RAN comprises a radio network controller (RNC) 125, which is responsible for radio resource management functions and certain network mobility management functions. By way of example, data network 103 is configured as an IP/IMS (IP Multimedia Subsystem) with multiple application servers 115 that supply multimedia content. Data network 103 couples to PSTN 105 through a media gateway 117; the PSTN can serve one or more 119 voice terminals.

[012] A Radio Access Bearer (RAB) is associated with the packet data protocol (PDP) context maintained between the User Terminal (UT) 111/114 and the Core Network (CN) 101. For example, a RAB can be established for Session Initiation Protocol (SIP) call signaling and be maintained as long as the user wants to make and receive calls. Another RAB is established on demand to transport voice media while a call is in session. The satellite radio access network establishes and maintains radio bearers (RBs) between the UT and the SBSS 107 necessary to satisfy, for example, the quality of service (QoS) requirements of SIP call signaling and Voice over IP (VoIP) user plane RABs. The signaling radio bearer supports the signaling connectivity between the UT and the satellite radio access network.

[013] The SIP protocol is typically used to establish session initiation and management. A SIP message mainly contains three sections that detail the session, timing, and media descriptions. A packet data protocol (PDP) context is created for each initiated session, which contains the desired characteristics of the specific session, including the type of PDP and the required QoS among other parameters. A PDP context can be seen as a set of information maintained by UT, GGSN and SGSN. It contains a PDP type that identifies the packet data network (PDN) type, the PDP address, QoS information and other session information. Enabling a PDP context refers to creating the PDP context in the UT, SGSN and GGSN such that the UT can communicate with an entity in PDN using the PDP address held in the PDP context. Additionally, a secondary PDP context activation allows the subscriber to establish a PDP context with a different QoS profile for the same PDN.

[014] The core network (CN) 101 may include a proxy call session control function (P-CSCF), a service call session control function (S-CSCF), an interrogation call session control function (l-CSCF), a media resource function controller (MRFC), a media resource function processor (MRFP), a media communication port (MGW), a media communication port controller (MGCF) function, and a signaling communication port (SGW). Note that these components are the components that refer to the Session Initiation Protocol (SIP). For other applications, however, the CN 101 may include different components. Additionally, all such components associated with SIP signaling are known in the art and, therefore, are not shown in the Figures and their functionality is not discussed in detail in the present invention. Furthermore, as will be appreciated, the system 100 can take many forms and include multiple components and facilities and/or alternative components and facilities.

[015] According to the exemplary embodiments, approaches are provided to perform an automatic terminal change, taking into account a multitude of factors for the selection of the best set of available trajectories. By way of example, a set of a satellite, satellite beam, and frequency for forward and reverse link directions need not be the same, and no symmetry is assumed in forward and reverse link direction beam patterns. Exemplary algorithms, in accordance with embodiments of the invention, can combine a set of metrics to select an optimal trajectory, balance multiple parameters such as signal-to-noise ratio, expected duration of beam coverage, and other parameters, described in further detail below. The approaches additionally provide continuous automatic switching from current trajectories to alternative trajectories considered more ideal, in which no loss of transmission data is experienced, except when an automatic change involves an automatic frequency change. As an additional example, rather than just considering channel condition as the primary factor in making automatic switching decisions, the example embodiments provide consideration of several relevant factors and a flexible method for weighting such factors based on, for example, system characteristics and system design objectives. In satellite communications systems, automatic switching comprises the reassignment or reconfiguration of radio links between a terminal and a satellite, a satellite beam and carrier frequency set or triplet. The criteria for performing an automatic switch vary from system to system, but in general, the automatic switches are triggered to select paths with better channel conditions and, therefore, only consider the channel conditions on both the current paths and potential available paths as the factors for making an automatic switch decision. Additionally, for use in the present invention, a "path" comprises a satellite, a satellite beam and carrier frequency triplet, and a "subbeam" refers to a pair of satellite beam and carrier subband. In line with additional exemplary embodiments, such approaches provide for consideration of an expanded set of factors to achieve a more effective and improved automatic change process. Such expanded factors, for example, are largely driven by system characteristics (eg, in a LEO system, the orbiting speed of the satellites, current and available transmit power levels, and available frequencies). Additionally, according to such embodiments, multiple paths can be selected to provide processing gain and/or diversity gain.

[016] Figure 2 illustrates a diagram representing exemplary beam patterns of two respective satellites, satellite 1 (SAT-1) and satellite 2 (SAT-2), which move in different directions, as reflected by the trajectory lines 201, 203. Each beam is identified with a letter of the alphabet and each beam can comprise multiple carrier frequencies. Shown is a user terminal (UT) 205 located in an overlapping section of the two beam patterns. The UT is situated in the overlapping area of satellite beams from Satellite 1, H-beam and Satellite 2, C-beam, and, for example, can be serviced using multiple (e.g., two) paths to provide processing gain and path diversity, however, the beams need to have the same operating carrier frequency. Alternatively (not shown in the Figure), adjacent beams from one satellite may overlap, and a terminal within the overlapping portion of two beams may be serviced by both beams simultaneously. In a GEO satellite communications system, where the satellite projection area on Earth is fixed, the UT may require automatic switching due to its mobility, changing channel conditions and underbeam coverage.

[017] According to exemplary embodiments of the present invention, in a mobile satellite communications system, several different types of automatic changes can be performed. According to one example, an automatic forward link switch can be initiated by the RAN such that it commands the UT to switch one or more of its downlink receive paths (forward link paths), and reconfigure the associated downlink transmission paths accordingly. By way of example, one type of automatic forward link switching is an assisted mobile automatic switching. In a mobile-assisted automatic switchover, the RAN can use forward path measurement reporting procedures to obtain feedback on the quality of target forward links. Assisted mobile automatic changes are always intra-frequency (automatic changes involving common carrier frequency paths), but may involve automatic inter-beam or inter-satellite path changes. An assisted mobile autochange can comprise a "soft" autochange - where a change of one or more active trajectories is performed while at least one active trajectory remains constant (when trajectory diversity is in effect), or a "hard" autochange, which involves a reconfiguration of all active trajectories at the same time. By way of further example, another type of automatic forwarding link switching is a hidden automatic switching. In a hidden automatic shift, no trajectory measurement is used. Hidden automatic shifts can be performed in situations when measurements are unavailable or impossible to obtain (for example, in interfrequency automatic shifts). Hidden autoshifts are usually (if not always) hard autoshifts.

[018] According to a further example, an automatic return link switch can be initiated by the RAN, so that it selects a different set of uplink receive paths (return link paths). By way of example, one type of reverse link automatic switching is an intra-frequency automatic switching (an automatic switching involving common carrier frequency paths). Automatic intra-frequency return link changes are transparent to the UT, and do not require message exchange between the RAN and the terminal. Automatic intra-frequency return link changes can be soft or hard. By way of further example, another type of automatic reverse link switching is an automatic inter-frequency switching (an automatic switching that involves changing carrier frequency paths). In an automatic inter-frequency reverse link switch, the RAN commands the UT to switch its respective uplink frequency.

[019] In line with the exemplary autoshift approaches provided in this document, trajectory metrics are used to evaluate trajectory options for different autochange scenarios. For example, a path selection metric (PSM) is a metric for a path, and a combined path metric (CPM) is a metric for a set of paths. Such metrics are used to evaluate existing radio links and to make decisions about the likelihood of performing an automatic switchover and the selection of the appropriate set of paths for the automatic switchover.

Path Selection Metric (PSM):

[020] According to the exemplary modalities, the trajectory selection metric (PSM) is used in the selection of the best candidate trajectory, and seeks to combine several factors for the evaluation of a subbeam in relation to others.

[021] As an example, one type of PSM consists of a normalized relative pilot quality/intensity metric (S). With pilot strength-quality PSM, when forward link pilot measurements are available (for example, when the UT is tuned on a channel and is able to measure certain characteristics of that channel), the relative pilot quality/strength metric uses a measurement of a pilot signal-to-interference ratio (SIR) or signal-to-noise ratio (S/N), plus a correction for differences in actual pilot transmit power between the forward link beams that are measured. The pilot is a reference signal transmitted by the RAN, which is always being transmitted at a consistent transmit power. For example, the UT can measure the SIR for the current channels or forwarding link paths for which the UT is tuned and transmit those measurements back to the RAN. Such measurements may be performed using any generally known method for measuring signal-to-interference ratios on such a mobile satellite communication path or channel. Based on the SIR measurements received from the UT, the RAN can then determine the relative pilot quality/intensity (S) as follows:

[022] For return link pilots, the relative pilot quality/intensity (S) is the measured pilot SIR (without correction), normalized. For the return link SIR measurements, the RAN performs such measurements directly based on the return link transmissions received from the UT. Such measurements may be performed relative to the UT's active return link paths and/or other paths of the same frequency within beams (from the same satellite or from a different satellite) covering the current position of the UT. Although the RAN may not be decoding such other inactive paths, the RAN can receive transmissions from the UT, and if an automatic switchover is initiated, then one or more of these inactive paths can become active. For return link pilots, the relative pilot quality/intensity (S) can be determined as follows:

[023] When forward or reverse link SIR measurements are not available for a candidate subbeam, but measurements are available for a different subbeam on the same satellite beam (e.g. in an automatic interfrequency shift - within the same beam but for a different frequency path - the UT would be unable to perform SIR measurements for a path of a different frequency), the measurements available for the different subbeam can be used as an approximation to the desired subbeam measurements. Similarly, when forward or backlink SIR measurements are not available for a candidate beam, but measurements are available for a different beam (e.g., a different beam from the same satellite, but potentially for a different frequency path), then the available underbeam measurements can be used as an approximation for the desired underbeam measurements. The approximations, however, need to be adjusted for beam gains and pilot transmit powers - for example, approximations can be determined as follows:

[024] Those situations, where forward or return link measurements are not available, may occur in automatic interfrequency changes and other hidden automatic changes.

[025] When return or forward SIR measurements are not available for a candidate satellite (again, for example, when measurements cannot be made due to different frequencies), but measurements are available for a beam on a different satellite, then the available underbeam measurements can again be used as an approximation for the desired underbeam measurements (for example, after adding a beam and path gain compensation due to the different distances between the UT and the two satellites, and (for example) a configurable deficiency - See , for example, Figure 7) - for example, as shown below:

[026] Those situations, where forward or return link SIR measurements are not available for a candidate satellite, may occur in automatic inter-frequency changes and other hidden automatic changes. The configurable disability comprises a flexible variable that provides additional control over the initiation of automatic changes (e.g., the disability can be manipulated based on past experience, simulation results, and desired operational and system objectives regarding automatic changes, such as frequency of automatic changes versus signal quality and resulting QoS).

[027] In the above equations (1) to (6): PPilot,beam,J is the J-beam RAN pilot transmit power (dBW), Gbeam,J is the serviceband beam antenna gain at the UT location for the J-beam (dB), Gpath,J = 147.55 — 20 log(F x d) — 240, which is the path loss of free space on the mobile link for the J-beam F is the carrier frequency service band (GHz), d is the slope range (km), Smin is the expected minimum Sadj is a configurable adjustment factor to account for losses and to serve as an adjustable handicap (eg -3 dB). Deficiency reflects an adjustable parameter to adjust estimates, for example, which can be determined based on past experience with the same system, simulations, design objectives in relation to frequency of automatic or similar changes (e.g., aggressiveness in relation to automatic change decision making), etc.

[028] As an additional example, another type of PSM consists of a subbeam lifetime (L) metric. The subbeam lifetime, relative to a UT, is the length of time that the subbeam is likely to remain a physically viable trajectory for the UT, which is derived based on the satellite trajectory and its beam coverage relative to the terminal position. For example:

[029] The value of Llow and Lhigh can be chosen to reflect the duration of beam coverage based on the particular satellite system (eg, for a LEO system, 10 and 30 seconds, respectively). For example, based on the system design (e.g. orbit and velocity of the satellites, and beam patterns), Llow and Lhigh can be estimated and additionally such estimates can be adapted to affect the auto-change frequency (e.g. pushing the estimates further towards the system limits will reduce the auto-change frequency, while more conservative estimates will result in a higher auto-change frequency). Furthermore, Llow and Lhigh normalize the anticipated beam coverage time across the system.

[030] As an additional example, another type of PSM consists of an auto-changing trend metric (H). The autoshift bias can be used to delay the autochange until a suitable (eg sufficiently better) trajectory is found. For example, if the subbeam is a current active trajectory, then H = 1, or if the subbeam is a candidate trajectory, then H = 0.

[031] As an additional example, another type of PSM consists of a metric that reflects an available capacity factor (or inverse load factor) for the subbeam (A). An available capacity factor (or reverse load factor) for the forward link subbeam, for example, might be as follows: where Pcommitted reflects the current committed power and Pmax is the maximum power allowed in a subbeam. Consequently, the greater the committed power relative to the maximum power, the more limited the capacity becomes - until Pcommitted = Pmax, and then A becomes zero. Additionally, an available capacity factor (or reverse load factor) for the reverse link subbeam, for example, may comprise that A is based on the uplink allocation/existing flows ratio and the maximum allowable channels, where more uplink transmissions or flows increase interference.

[032] According to one modality, based on the exemplifying trajectory selection metrics mentioned above, the PSM can then be calculated as follows: where Ks, KL, KH and KA provide different weights for different metrics and flexibility of system operation. As an example, the weights can each be defined according to a weight to be assigned to the respective metric to address the various system tradeoffs (such as weighting of pilot measurements versus underbeam lifetime). In this regard, a larger weight applied to the relative pilot quality/intensity (S) may lead to a higher frequency of automatic changes (e.g. a better quality trajectory may then trigger an automatic change even though there is a relatively long subbeam lifetime), whereas a larger weight applied to the subbeam lifetime (e.g. a better quality trajectory may then not trigger an automatic change when there is a relatively long subbeam lifetime). Consequently, with respect to PSM determination, the embodiments of the present invention provide a generic approach that can be easily adapted to realize different system and design goals and efficiencies (e.g., based on the various operational tradeoffs of the system - such as minimizing automatic changes by sacrificing signal quality at the margins versus maximizing signal quality on a consistent basis by sacrificing the overhead associated with a higher frequency of automatic changes).

Combined Path Metric (CPM):

[033] According to the exemplary embodiments, the combined trajectory metric (CPM) is used in selecting a set of candidate trajectories that provide good trajectory diversity. A set of trajectories whose CPM is greater than that of another set is considered as a better choice for a set of diversity trajectories. As an example, a CPM can be determined as follows. First, the PSM(i) = the PSM of paths i, i = 1, ... , n (the PSM of each individual path i) is calculated as described above. Then, a trajectory diversity parameter D(1, ... , n) for the set of trajectories 1, ... , n is selected based on the diversity level of each trajectory (e.g., where the diversity parameters favor a high degree of trajectory diversity - e.g., four satellites are better than 3, 2, or 1, for example, due to increased line-of-sight options). The path diversity parameters, for example, can be selected as follows: 0.0 if n = 1, 0.1 if n > 1 and paths 1, ..., n are on one satellite, 0.5 if n > 1 and paths 1, ..., n are on two satellites, 0.8 if n > 1 and paths 1, ..., n are on three satellites, 1.0 if n > 1 and paths 1, ..., n are on four satellites .

[034] Then, the CPM of the set of trajectories 1, n can be determined as follows: where KD is a "diversity coefficient" (eg 3 dB) and thus (KD * D (1, ... , n)) creates a bias that favors increased satellite diversity. The diversity factor D (l, ... , n) thus reflects a flexible factor that can be defined to meet the system design objectives.

Selection of trajectories using CPM:

[035] According to the exemplary embodiments, the following algorithm can be used to select the best set of trajectories from a candidate set of subbeams using CPM, given a target number of NP trajectories and NS satellites of diversity. Optionally, the TF target frequency can be suppressed. First, the candidate list of subbeams is sorted by decreasing PSM. Then, separate the sets of candidate trajectories for evaluation and select the optimal set based on the result of the evaluation, for example, as follows: (1) Separate the first subbeams N < NP so that (a) each subbeam frequency is TF, (b) the SIR Si of each subbeam (as used in the PSM computation) is within ΔSmax of that of the primary trajectory, and (c) the candidate list spans different NS satellites. Request this candidate set C(NS) and compute CPM(C(NS)) - if the NS satellites are not found, the candidate set C(NS) is therefore empty and not considered; (2) Decrease NS and repeat (1) until candidate sets C(NS), C(NS— 1), ... , C(l) have been generated; and (3) select the ideal candidate set as the set C(i) with the highest CPM value.

Auto-changing forward path selection:

[036] According to the exemplary modalities, the RAN periodically re-evaluates the set of forward path candidates and active paths. For example, the RAN may re-evaluate the set of forward path candidates and active paths based on a predetermined periodicity (e.g., where the periodicity reflects another flexible factor used to achieve desired system performance and operation - e.g. every 10 seconds). As a further example, the RAN may re-evaluate the set of forward path candidates and active paths when certain events occur that may trigger an automatic switch (for example, such events include UT position updates, measurement reports, and certain resource management events - such as a resource management action that will make a beam no longer available for a particular UT).

[037] According to the additional exemplary arrangements, special criteria may apply when certain automatic trajectory changes are being considered. By way of example, one such criterion might be the fact that the selected carrier frequency needs to be equal to other paths in the diversity set that are not being delivered. In the case of an automatic interfrequency change, all diversity paths are delivered together. By way of further example, another such criterion for an automatic inter-frequency change could be the fact that (if reasonably possible) the target trajectory needs to be on the same satellite as the existing trajectory, which would allow the UT to use approximations based on current measurements to acquire the target pilots. As an additional example, such an additional criterion for an automatic intersatellite switch could be the fact that the target carrier frequency needs to be the same as the current one, if possible. This allows the UT to take measurements of the target pilots before the automatic switch.

[038] According to one embodiment, the process of automatically changing the forwarding path is as follows:

[039] STEP I: The automatic switching process tries to keep the active paths at the same F1 frequency. Based on this goal of keeping the active trajectories at the same F1 frequency, first determine the list of candidates at the current active F1 frequency based on the last pilot measurements reported by the UT and resulting computed PSMs for all trajectories. Then, the algorithm described above in the section entitled "Selecting trajectories using CPM" is applied to select the optimal set of trajectories, for example, using certain options in the algorithm, as follows: Target frequency TF = F1 (the current frequency); Target number of paths NP = 2 and Target number of satellites NS = 2. The number of paths and number of satellites in this example are each 2, however (as would be evident) other values could be used - for example, if no diversity was desired or required due to system constraints or number of receivers at the terminal, the NP and NS could be set to 1. independent data that a UT may technically be able to receive). Finally, the determined ideal set is examined, such that: (1) If the winners are equal to the current active trajectories, no automatic change is required in that frequency; (2) If the winners are not equal to the current active trajectories, then the combined trajectory metric (CPM) of the winners is examined; (3) If the CPM of the winners is greater than the CPM of the current active trajectories, then automatic switching is triggered (i.e., this can be an automatic smooth intra-frequency change if a winner is an active trajectory, or an automatic hard intra-frequency change if no winner was an active trajectory); and (4) If the CPM of the winning set is less than that of the active set, no automatic change is required.

[040] STEP 11: In cases where the previous step did not recommend an automatic change in the current frequency (exams (1) and (4), from STEP I, immediately above), the active trajectories should be examined to see if an automatic change of frequency is required, as explained below. If the lifetime of each of the current active trajectories L > Lhigh seconds, then no automatic changes are required. Otherwise, a list is created of all subbeams in "release pending" or "active" state on the current active trajectory only satellite (or satellites), and their PSMs are computed (including the current active trajectories in that list). Where no measurements are available for the other frequencies, approximations can be used in PSM calculations. Then, apply the algorithm described above in the section entitled "Selecting trajectories using CPM" to select the optimal set of trajectories, for example, using certain options in the algorithm, as follows: Target frequency TF = unspecified - the algorithm separates the frequency of the top candidate; Target number of trajectories NP = 1; and Target number of satellites NS = 1. The number of tracks and the number of satellites in this example are each 1, however (as would be evident) other values can be used. Additionally, the values of NP and NS may be constrained by system characteristics (for example, limitations on the number of independent path transmissions that a UT may technically be capable of transmitting). Additionally, in this example, the UT may only be able to transmit on one path, but the RAN may receive that transmission on all paths of the same frequency and beams covering the UT's position. Finally, the determined ideal set is examined, so that: (1) If the winners have a CPM greater than the CPM of the current active trajectories, then an automatic interfrequency change is performed on the same satellite (or satellites); (2) If a better set of trajectories has not been found and the lifetime of each of the current active trajectories is L > Llow seconds, then no automatic change is required; (3) Otherwise, an automatic intersatellite switch should be considered - repeat the above steps of creating a list of all subbeams in "active" or "release pending" state on the current active track satellite (or satellites), and compute their PSMs (including the current active tracks in that list), but including subbeams from other satellites in the list; (4) If a better set of trajectories is found, then an automatic interfrequency switch to a different satellite is performed; and (5) If no better trajectory is found, then no automatic change is performed.

Automatic shifting return path selection:

[041] According to the exemplary embodiments, at the time of automatic switching, the RAN generally has measurements of multiple candidate paths provided by pilot surveys during the course of the respective data session or active call. Additionally, with an active call or data session, an active traffic channel frequency has been established, so the RAN tries to keep the UT on the same return frequency during automatic changes.

[042] According to one embodiment, the RAN uses the following algorithm to choose the auto-changing return paths:

[043] STEP I: On return path selection, the auto changeover process tries again to maintain the current active traffic frequency F1. Based on this goal of maintaining the same F1 active frequency, first determine the candidate list of all return subbeams in the UT beam path and rank them by decreasing PSM (where the actual measured pilot SIRs are used to calculate the PSMs). Then, the algorithm described above in the section entitled "Selecting trajectories using CPM" is applied to select the optimal set of trajectories, for example, using certain options in the algorithm, as follows: Target frequency TF = F1 (the current frequency); Target number of trajectories NP = 2; and Target number of satellites NS = 2. The number of tracks and number of satellites in this example are each 2, however (as would be evident) other values could be used - for example, if additional diversity was required or desired to optimize diversity gain or processing gain, then the NP and NS could be set to 4. This number could also depend on the number of satellites visible to the UT and the size of the satellite constellation. Finally, the determined ideal set is examined, such that: (1) If the set of winners equals the current active set of return paths, then no automatic change is required at that frequency; (2) If the set of winners differs from the current active set, then an intra-frequency backtrack reconfiguration is performed, which may involve adding, replacing, or removing one or more active tracks.

[044] STEP 11: In cases where the previous step did not recommend an automatic change in the current frequency F1(exam (1), from STEP I, immediately above), the active paths should be examined to see if an automatic change of frequency is required, as explained below. If the lifetime of at least two of the current active trajectories is L > Lhigh seconds, then no automatic change is required. This is done due to the fact that the path selection can be tried again, for example after 10 seconds, at which time better candidates could become available. Otherwise, a list is created of all subbeams on the current active track only satellite (or satellites) and computed their PSMs (including the current active tracks in that list). Where no measurements are available for the other frequencies, measurements at the current active frequency can be used as an approximation in PSM calculations. Then, apply the algorithm described above in the section entitled "Selecting trajectories using CPM" to select the optimal set of trajectories, for example, using certain options in the algorithm, as follows: Target frequency TF = unspecified - the algorithm separates the frequency of the top candidate; Target number of trajectories NP = 2; and Target number of satellites NS = 2. The number of tracks and number of satellites in this example are each 2, however (as would be evident) other values could be used - for example, if additional diversity was required or desired to optimize diversity gain or processing gain, then the NP and NS could be set to 4. This number could also depend on the number of satellites visible to the terminal and the size of the satellite constellation. Finally, the determined ideal set is examined, so that: (1) If the winners have a CPM greater than the CPM of the current active trajectories, then an automatic interfrequency return switch is performed on the same satellite (or satellites); (2) If a better set of trajectories has not been found and the lifetime of at least two of the currently active trajectories is L > Llow seconds, then no automatic change is required at that time; (3) Otherwise, an automatic intersatellite switch should be considered - repeat the above steps of creating a list of all subbeams on the current active track satellite (or satellites) and compute their PSMs (including the current active tracks in that list), but including subbeams from other satellites in the list; (4) If a better set of trajectories is found, then an automatic inter-frequency return switch to a different satellite is performed; and (5) If no better trajectory is found, then no automatic change is performed.

Automatic switching procedure in CDMA:

[045] According to additional exemplary embodiments, processes for automatic switching in a CDMA system are provided. Figure 3 illustrates a signal timing diagram representing a process for physical channel reconfiguration of frequencies or forwarding paths, for an automatic user terminal (UT) change situation in a CDMA wireless communication system, in accordance with exemplary embodiments of the present invention. By way of example, for an automatic forward link switch, the RAN provides certain information to instruct the UT to perform the automatic switch (S301), including the satellite identification, beam and carrier frequency, the Walsh codes, and the switch time (e.g., specifying the forward frame number at which the RAN and terminal switch to the new paths). The satellite and beam information provides the proper mapping for a pseudo-random noise (PN) code to be used by the terminal. When forward path diversity is being used, the reconfiguration can include multipath information. The frequency is, in general, the same for all trajectories. With diversity in use, a soft autoshift involves reconfiguring some trajectories while keeping at least one trajectory active, whereas a hard autoshift involves reconfiguring all trajectories active. The RAN continues to transmit data on the old paths until the target toggle frame (S303). The UT reconfigures the receiver (S305) and the RAN reconfigures the transmitter (S307). Depending on the type of reconfiguration (for example, downlink frequency change), the RAN may pause downlink traffic for a short period during the transition to allow the UT reconfiguration to be activated. Then, in the toggle frame, the UT activates the new configuration (S309) and the RAN activates the new configuration (S311). The UT then sends a message to the RAN indicating that the physical channel reconfiguration is complete (S313), and the RAN continues to send data (after the toggle frame) based on the new configuration (S315).

[046] Figure 4 illustrates a signal timing diagram that represents a process for reconfiguration of the physical channel of frequencies or return paths, for an automatic change of user terminal (UT) situation in a CDMA wireless communication system, according to the exemplary embodiments of the present invention. By way of further example, for an automatic switchback, the RAN provides information to instruct the UT to perform the automatic switchover (S401). UT transmissions on the assigned return frequency are received by the RAN through several paths (satellites and beams). Active RAN receivers and path diversity are transparent to the user. Additionally, Walsh codes to be used on physical return channels can be included in the PCR/RBR message, if they need to be changed. There is normally no need to change return channel Walsh code assignments due to the fact that they are independent of the return subbeam. The UT continues to transmit data on the old paths until the target frame (S403). The UT reconfigures the transmitter (S405) and the RAN reconfigures the receiver (S407). Then, in the toggle frame, the UT activates the new configuration (S409) and the RAN activates the new configuration (S411). The UT then sends an initial pilot only transmission (S413) and then sends a message to the RAN indicating that the physical channel reconfiguration is complete (S315). The UT then continues to send data (after the toggle frame) based on the new setting (S417).

TDMA System Automatic Switching Procedure

[047] According to additional exemplary embodiments, processes for automatic switching in a TDMA system are provided. In a TDMA system, the return and forward beam patterns are assumed to be identical and overlapping. Figure 5 illustrates a signal timing diagram that represents a process for physical channel reconfiguration of frequencies or return and forwarding paths, for an automatic user terminal (UT) change situation in a TDMA wireless communication system, in accordance with exemplary embodiments of the present invention. For system operation, each forwarding sub-beam has a corresponding uplink sub-beam, so the automatic switching applies to both the return and forwarding frequencies simultaneously. As an example, to instruct the UT to perform an automatic switchover, the RAN provides certain information to the UT via the radio bearer (RB) reconfiguration message (S503), including the forward carrier frequency, return carrier frequency, and switch time (e.g., specifying the forward link frame number on which the RAN and UT switch to the new paths). Prior to the RB reconfiguration message, the RAN computes a switching time (eg downlink activation time = N) for the reconfiguration (S501). This establishes a point in time reference to the downlink frame number base. For example, the target downlink frame number N is chosen such that sufficient time is allowed for the RAN to retransmit the RB reconfiguration message, and the uplink activation frame is based on the satellite hop delay. The RAN continues to transmit and receive voice packets on the old channels until downlink (DL) frame N is reached, but (S501) the RAN stops transmitting and receiving data traffic for all other flows that rely on segmentation information and sequence number to reconstruct upper layer packet data units (PDUs). The last successfully received sequence numbers that resulted in a complete upper layer packet assembly are also sent to the UT in the reconfiguration message in order to keep the RAN and UT segmentation/reassembly states in sync. When the UT receives the reconfiguration command from the RB, it validates the configuration and stops all data traffic flows up to the DL frame N (except the voice flow, if present) (S505). When the DL frame N is reduced in the RAN, it reconfigures the physical downlink channel according to the new configuration and starts transmitting bursts on the new DL channel (S507). At that point, the RAN can start transmitting voice packets on the new downlink channel (S509), space and reconfigure the uplink channel (S511). Similarly in DL frame N, the UT reconfigures both the downlink and uplink physical channels and starts transmission on the new uplink physical channel (S513). At that point, the UT continues transmission of voice packets (if present) on the new uplink channel (S515). The UT then transmits an RB reconfiguration completion message to the RAN using the new uplink channel (S517). When the RB reconfiguration completion message is received at the RAN, the RAN acknowledges the message (S519) and resets all other data traffic flows (S521). Finally, the UT re-establishes all other data traffic flows (S523).

TDMA system automatic satellite switching

[048] In TDMA systems, all UTs need to be synchronized so that their transmissions are properly aligned on the satellite and do not overlap in time. In an exemplary embodiment, the synchronization is based on a nominal point on the Earth's surface, for example, the center of the beam. Figure 6 illustrates an uplink allocation timing diagram for transmissions relative to nominal positions, for an automatic user terminal (UT) switch situation in a TDMA wireless communications system, in accordance with exemplary embodiments of the present invention. Figure 7 illustrates a diagram representing transmission propagation delays for an automatic user terminal (UT) switching situation in a TDMA wireless communication system, in accordance with exemplary embodiments of the present invention. Referring to Figure 6, timing proceeds so that all transmissions from the UTs arrive aligned to the satellite. As an example, assume that TNominalProp is the transmission propagation time between the satellite and the nominal point and that AllocationToTxDelay is the time between the receipt of an uplink allocation and the uplink transmission to a UT at the nominal point. Thus, a UT at a P1 position, with transmission propagation delay from TUTProp to the satellite, will define its AllocationToTxDelay(UT1) as AllocationToTxDelay(UT1) = AllocationToTxDelay — 2 * ( TUTProp — TNominalProp).

[049] Both AllocationToTxDelay and (TUTProp — TNominalProp) are normally provided by the RAN to the UT during the initial system access using the RACH procedure, where (TUTProp - TNominalProp) is calculated using the UT location sent in the RACH or using the RACH arrival time and comparing with the expected RACH arrival time from the nominal location. When a UT needs an automatic switch to a different satellite, the timing on the UT needs to be updated to reflect the UT to target satellite propagation delay, the target satellite's nominal point on Earth, and its AllocationToTxDelay value. The AllocationToTxDelay parameter is normally different for each set of satellites/beams. In order to achieve this, the RAN uses a UT position and the satellite ephemeris data to calculate the following parameters, which are sent in an autochange message along with the new return and forward frequencies: minalPropagationDelay(Target) (14)

[050] Based on TxPropagationDelayUpdate, UT can synchronize with new downlink traffic channel without using downlink frequency channel, and based on PropagationDelayDiffWithNominal(Target), UT can update its AllocationToTxDelay value to AllocationToTxDelay(Target) - 2 * PropagationDelayDiffWithNominal(Target) - See for example Figure 7.

[051] Figure 8 illustrates a signal timing diagram for an intersatellite automatic switching process, for an automatic user terminal (UT) switching situation in a TDMA wireless communications system, in accordance with exemplary embodiments of the present invention. Referring to Figure 8, the flow reflects an automatic change between two satellites (automatic change from satellite 1 to satellite 2). For example, after the initial auto switch instruction (S801), the UT reconfigures its receiver (switching to the new frequencies) and decodes the downlink traffic channel on the new frequency (S803). Using the newly adjusted timing correction (eg, AllocationToTxDelay), the UT transmits an uplink PRACH to access the system using the traffic channel (S805). PRACH allocations are signaled in downlink transmission similar to the allocation of uplink data transmission. The initial access with PRACH allows the RAN to instruct the UT to do additional timing adjustment to correct residual timing errors. The PRACH also carries an auto-change message instruction reference number to assist the RAN in locating the UT context. After the UT receives the flow assignment from the RAN (S807), the UT sends a physical channel completion message and reactivates all flows that have undergone automatic switching (S809). Using PRACH has a less stringent synchronization and timing requirement than using a traffic channel. If, however, the RAN deems that the timing accuracy is sufficient for the UT to use the traffic channel immediately, then the RAN will instruct the UT not to send a PRACH first and to immediately use the traffic channel and listen to the uplink allocation to transmit the physical channel completion message. Then, data transmissions continue on the downlink and uplink over satellite 2 (S811).

Auto changeover with radio network controller (RNC) relocation:

[052] According to the exemplary embodiments, an automatic switching with RNC relocation process is used to perform continuous automatic switching of a UT and active UT sessions from one radio network controller (RNC) to another. Figure 9 illustrates a signal timing diagram representing an inter-RNC automatic switching process, for a user terminal (UT) automatic switching situation in a wireless communication system, in accordance with exemplary embodiments of the present invention. The RNC is a governing element in a Universal Mobile Telecommunications System (UMTS) radio access network and is generally responsible for controlling the connected to serve the respective UTs. The RNC generally performs radio resource management functions, some mobility management functions, and is the encryption point for user data traffic. By way of example, automatic inter-RNC switching is triggered by fixed UT position knowledge or by position updates or neighboring beam measurements sent by the UT from which the RAN determines that the UT has crossed into the target beam.

[053] According to an exemplary modality, the process is carried out as follows:

[054] First, the relocation preparation is carried out. By way of example, for relocation preparation, the originating RNC sends a Radio Access Network Application Part (RANAP) Relocation Required message to the SGSN containing the necessary context information for all protocol layers from the Radio Access Bearer (RAB) information to the physical layer (S901). The SGSN augments this with the RAB information available in the core network and sends this in the relocation request message to the target RNC (S903), for example, which may include: UT identification, access network (NAS) stratum mobility and routing information, and RAB parameters (e.g., QoS and GTP-U information). The relocation required message generated by the source RNC contains a "transparent container from source RNC to target RNC", which should be sufficient for the target RNC to allocate the physical layer resources and reconstruct the upper layer state and configuration for automatic changeover. The transparent container from source RNC to target RNC, for example, includes: a chosen algorithm, integrity protection key and state information, a chosen algorithm, encryption state and key information, UT radio access capabilities and UT type, UT position, and radio bearer (RB) and RAB information (RAB Id, RB Id, packet data convergence protocol (PDCP) configuration information, radio link control (RLC) configuration). Upon receipt of the RANAP relocation request message, the target RNC performs admission control and resource allocation for the RABs that are delivered (sets the RABs), and sends a relocation request acknowledgment message back to the SGSN (S905). With the physical channels assigned to the UT in the autoswitch, the target gateway (GW) constructs an RB reconfiguration message that contains the complete configuration required to perform the autoswitch, and embeds it within a "transparent container from target RNC to source RNC" in the relocation request acknowledgment sent to the SGSN, which, for example, includes: RB, PDCP information, new NAS and AS mobility information and identifiers (GRA, RA, U-RNTI), and main BCCH SI parameters for the target beam. The SGSN forwards the autochange command from the target RNC to the source RNC, embedding it in the relocation command message (S907).

[055] Second, the automatic changeover execution is performed. By way of example, for performing the autoswitch, the originating RNC triggers the autoswitching process by stopping all non-voice traffic flows and sending the autoswitching command (the RB reconfiguration message received from the target RNC via the CN) to the UT (S909). Voice traffic flow continues until the target downlink (DL) frame number N is reached. The originating RNC computes a switching time (eg, downlink activation time = N) for the reconfiguration, which establishes a reference time point based on the DL frame number. The target DL frame number N is chosen so that sufficient time is allowed for the RAN to retransmit the RB reconfiguration message. The Uplink Activation (UL) frame is based on the satellite hop delay. The RNC continues to transmit and receive voice packets on the old channels until the DL frame N is reached. The RNC stops transmitting and receiving data traffic for all other flows that rely on segmentation and sequence number information to reconstruct upper layer packet data units (PDUs). The last successfully received sequence numbers that resulted in a complete upper layer packet assembly are also sent to the UT in the reconfiguration message in order to keep the RAN and UT segmentation/reassembly states in sync. When the UT receives the reconfiguration command from the RB, it stops all data traffic flows up to the downlink frame N, except the voice flow, if present. Queued non-voice data that is received is forwarded towards the target RNC (S911). This data will be queued at the target RNC until different voice streams are re-established (for example, the end of the reconfiguration procedure). In the N activation frame, the voice packets are also forwarded to the target RNC (S913) to be sent on the newly configured channel. The CN starts forwarding data to the target RNCS at a later stage, when the reallocation is detected and completed. Thus, by forwarding voice traffic in the wake-up frame as well, minimal voice traffic interruption is experienced by the UT.

[056] When the DL frame N is reached at the target RNC, it reconfigures the downlink physical channel according to the new configuration and starts transmitting voice traffic on the new downlink channel to the UT. At that point, the voice packets are transmitted on the new DL channel (S915). Similarly in DL frame N, the UT reconfigures both the downlink and uplink physical channels and starts transmission on the new uplink physical channel (S917). At that point, the UT continues to transmit voice packets, if present, on the new uplink channel to the target RNC. Based on the downlink activation frame and the uplink activation time in the UT, the target RNC configures the new uplink physical channel and starts acquiring the transmission of uplink voice packets from the UT. The UT then transmits an RB reconfiguration completion message to the target RNC using the new uplink channel (S919). When the RB reconfiguration completion message is received at the target RNC, the target RNC acknowledges the message and re-establishes all downlink active data traffic flows (S921). When the target RNC receives the RB reconfiguration completion message from the UT, the UT acknowledges and resets all other data traffic flows on the DL. The target RNC sends the relocation completion (S925) and relocation detection (S923) messages to the SGSN, and the SGSN then switches user data plane GTP traffic to the target RNC. When the UT receives Layer 2 acknowledgment of the RB reconfiguration completion message, the UT re-establishes all other data traffic flows on the uplink.

Automatic intersatellite switching with radio network controller (RNC) relocation:

[057] According to the exemplary embodiments, an intersatellite automatic switch with the RNC relocation process is used for the automatic switch of a UT from a current satellite (source satellite) to a new satellite (target satellite), and the automatic switch of the UT and active UT sessions from a current RNC (source RNC) to another RNS (target RNC). Figure 10 illustrates a signal timing diagram representing an intersatellite and inter-RNC automatic switching process, for an automatic user terminal (UT) switching situation in a wireless communications system, in accordance with exemplary embodiments of the present invention. The Figure shows the interaction between the various entities and over which satellite the messages are exchanged. This automatic switching process operates virtually the same as the process detailed above for an automatic switching with radio network controller (RNC) relocation, with respect to Figure 9, except that certain data and message transmissions are retransmitted either on the source satellite or on the target satellite, such satellite retransmissions are reflected by the signal arrows in Figure 10 (i.e., where a data transmission signal or message arrow points to one of the satellites and then continues to the destination, that data transmission signal or message message is being retransmitted on the respective satellite).

[058] According to an exemplary modality, the process is carried out as follows:

[059] First, the relocation preparation is carried out. By way of example, for relocation preparation, the originating RNC sends a Radio Access Network Application Part (RANAP) Relocation Required message to the SGSN containing the context information needed for all protocol layers from the Radio Access Bearer (RAB) information to the physical layer (S1001). The SGSN augments this with the RAB information available in the core network and sends this in the relocation request message to the target RNC (S1003), for example, which may include: UT identification, access network (NAS) stratum mobility and routing information, and RAB parameters (e.g., QoS and GTP-U information). The relocation required message generated by the source RNC contains a "transparent container from source RNC to target RNC", which should be sufficient for the target RNC to allocate the physical layer resources and reconstruct the upper layer state and configuration for automatic changeover. The transparent container from source RNC to target RNC, for example, includes: a chosen algorithm, integrity protection key and state information, a chosen algorithm, encryption state and key information, UT radio access capabilities and UT type, UT position, and radio bearer (RB) and RAB information (RAB Id, RB Id, packet data convergence protocol (PDCP) configuration information, radio link control (RLC) configuration). Upon receipt of the RANAP relocation request message, the target RNC performs admission control and resource allocation for the RABs that are delivered (sets the RABs), and sends a relocation request acknowledgment message back to the SGSN (S1005). With the physical channels assigned to the UT in the autoswitch, the target gateway (GW) constructs an RB reconfiguration message that contains the complete configuration required to perform the autoswitch, and embeds it within a "transparent container from target RNC to source RNC" in the relocation request acknowledgment sent to the SGSN, which, for example, includes: RB, PDCP information, new NAS and AS mobility information and identifiers (GRA, RA, U-RNTI), and main BCCH SI parameters for the target beam. The SGSN forwards the autochange command from the target RNC to the source RNC, embedding it in the relocation command message (S1007).

[060] Second, the automatic changeover execution is performed. By way of example, for performing the autoswitch, the originating RNC triggers the autoswitching process by stopping all non-voice traffic flows and sending the autoswitching command (the RB reconfiguration message received from the target RNC via the CN) to the UT (S1009) - which is relayed via the originating satellite. Voice traffic flow continues until the target downlink (DL) frame number N is reached. The originating RNC computes a switching time (eg, downlink activation time = N) for the reconfiguration, which establishes a reference time point based on the DL frame number. The target DL frame number N is chosen so that sufficient time is allowed for the RAN to retransmit the RB reconfiguration message. The Uplink Activation (UL) frame is based on the satellite hop delay. The RNC continues to transmit and receive voice packets on the old channels until the DL frame N is reached. The RNC stops transmitting and receiving data traffic for all other flows that rely on segmentation and sequence number information to reconstruct upper layer packet data units (PDUs). The last successfully received sequence numbers that resulted in a complete upper layer packet assembly are also sent to the UT in the reconfiguration message in order to keep the RAN and UT segmentation/reassembly states in sync. When the UT receives the reconfiguration command from the RB, it stops all data traffic flows up to the downlink frame N, except the voice flow, if present. Queued non-voice data that is received is forwarded towards the target RNC (S1011). This data will be queued at the target RNC until different voice streams are re-established (for example, the end of the reconfiguration procedure). In the activation frame N, the voice packets are also forwarded to the target RNC (S1013) to be sent on the newly configured channel. The CN starts forwarding data to the target RNCS at a later stage, when the reallocation is detected and completed. Thus, by forwarding voice traffic in the wake-up frame as well, minimal voice traffic interruption is experienced by the UT.

[061] When the DL frame N is reached at the target RNC, it reconfigures the downlink physical channel according to the new configuration and starts transmitting voice traffic on the new downlink channel to the UT. At that point, voice packets are transmitted on the new DL channel (S1015) - which are retransmitted on the target satellite. Similarly in DL frame N, the UT reconfigures both the downlink and uplink physical channels and starts transmission on the new uplink physical channel (S1017) - which is retransmitted on the target satellite. At that point, the UT continues to transmit voice packets, if present, on the new uplink channel to the target RNC. Based on the downlink activation frame and the uplink activation time in the UT, the target RNC configures the new uplink physical channel and starts acquiring the transmission of uplink voice packets from the UT. The UT then transmits an RB reconfiguration completion message to the target RNC, using the new uplink channel (S1019), retransmitted on the target satellite. When the RB reconfiguration completion message is received at the target RNC, it acknowledges the message and re-establishes all active data traffic flows on the downlink (S1021) - which is retransmitted on the target satellite. When the target RNC receives the RB reconfiguration completion message from the UT, the UT acknowledges and resets all other data traffic flows on the DL. The target RNC sends the relocation completion (S1025) and relocation detection (S1023) messages to the SGSN, and the SGSN then switches user data plane GTP traffic to the target RNC. When the UT receives Layer 2 acknowledgment of the RB reconfiguration completion message, the UT re-establishes all other data traffic flows on the uplink.

[062] In the descriptive report mentioned above, several modalities were described with reference to the attached drawings. It will be evident, however, that several modifications can be made therein, and additional modalities can be implemented, without departing from the broader scope of the invention, as presented in the following claims. The specification and drawings should therefore be considered in an illustrative rather than a restrictive sense.

Claims (10)

1. Method, characterized in that it comprises: determining a set of path factors with respect to each of a plurality of communications paths comprising a satellite (109), a satellite beam and carrier frequency for a mobile user terminal (UT) (111, 114) in a mobile satellite communications system by a radio network controller (125); determining a path selection metric (PSM) for each communications path by the radio network controller (125), wherein the path selection metric (PSM) for each communications path is determined through a weighted calculation based on the respective set of path factors for the communications path; and determining whether to perform an automatic change of the user terminal (UT) (111, 114) from a first path among the communications paths to a second path among a plurality of communications paths by the radio network controller (125), wherein the determination is based on an evaluation performed based at least in part on the path selection metrics (PSM) determined for each of the plurality of communications paths, wherein the forward and return link changes are evaluated independently, and wherein the set of path factors with respect to each of the communications paths comprises two or more aspects of the communications path. 2. Method, according to claim 1, characterized in that the aspects of the communications path comprise one or more indicators that each reflect a quality of the communications path, an expected lifetime of the communications path, an automatic change trend and a capacity factor. 3. Method, according to claim 2, characterized in that the one or more indicators that each reflect a quality of the communications path comprise a pilot intensity relative to the path, the expected service life of the communications path comprises an expected service life of a subbeam of the communications path, the automatic switching trend refers to the timing of the automatic switching, and the capacity factor reflects a potential transmission capacity of the communications path. 4. Method, according to claim 1, characterized in that the weighted calculation for determining the path selection metric (PSM) for each of the communications paths is computed based on a respective different weighting factor applied to each path factor for the communications path, and each different weighting is defined based on system characteristics and system design objectives. 5. Method according to claim 1, further comprising: determining a combined path metric (CPM) for each of a set of communications paths, wherein the set of communications paths comprises communications paths from the plurality of communications paths to which the user terminal (111, 114) can be delivered, wherein the combined path metric (CPM) is determined based at least in part on the path selection metric (PSM) for each communications path in the set of communications paths; and wherein each combined path metric (CPM) is computed to serve as an indicator reflecting a quality of path diversity with respect to the set of communications paths over which the user terminal can be delivered. 6. Access node (113) in a mobile wireless communications network, the access node (113) characterized in that it comprises: a receive interface configured to receive data communications from a mobile terminal (111, 114) over a plurality of network communications paths; a transmission interface configured to provide data communications for transmission to the mobile terminal over the network's plurality of communications paths; and a radio network controller (125) configured to (i) determine a set of path factors with respect to each of the plurality of communications paths, (ii) determine a path selection metric (PSM) for each communications path, wherein the path selection metric (PSM) for each communications path is determined through a weighted calculation based on the respective set of path factors for the communications path, and (iii) determine the probability of performing an automatic mobile terminal switch from a first path wherein the determination is based, at least in part, on the path selection metric (PSM) determined for each of the plurality of communications paths, wherein the forward and reverse link changes are evaluated independently, wherein each of the plurality of communications paths comprises a satellite (109), a satellite beam, and carrier frequency, and wherein the set of path factors with respect to each of the Communications Path comprises two or more aspects of the Communications Path. 7. Access node, according to claim 6, characterized in that the aspects of the communications paths comprise one or more indicators that each reflect a quality of the communications path, an expected lifetime of the communications path, an automatic change trend and a capacity factor. 8. Access node, according to claims 6 or 7, characterized in that the one or more indicators that each reflect a quality of the communications path comprise a pilot intensity relative to the path, the expected service life of the communications path comprises an expected service life of a subbeam of the communications path, the automatic changeover trend refers to the timing of the automatic changeover, and the capacity factor reflects a potential transmission capacity of the communications path. 9. Access node, according to any one of claims 6 to 8, characterized in that the weighted calculation for determining the path selection metric (PSM) for each of the communications paths is computed based on a respective different weighting factor applied to each path factor for the communications path, and each different weighting is defined based on system characteristics and system design objectives. 10. Access node according to any one of claims 6 to 9, characterized in that the radio network controller is further configured to: determine a combined path metric (CPM) for each of a set of communications paths, wherein the set of communications paths comprises communications paths from the plurality of communications paths for which the user terminal (111, 114) can be delivered, wherein the combined path metric (CPM) is determined based on the least in part on the path selection metric (PSM) for each communications path in the set of communications paths; and wherein each combined path metric (CPM) is computed to serve as an indicator reflecting a quality of path diversity with respect to the set of communications paths over which the user terminal can be delivered.