Satellite communication system with peer-assisted coverage extension

Abstract

The present disclosure provides a method for satellite communication comprising: transmitting, by an endpoint, a set of protocol data units (PDUs) at different spreading factors with a session request bit; receiving, by a satellite, a subset of the PDUs based on link conditions; constructing, by the satellite, a hash based on an endpoint ID and a minimum spreading factor of received PDUs; packing the hash with other data in a feedback channel; monitoring, by the endpoint, the feedback channel for a matching hash; if a matching hash is found, continuing transmission at the signaled spreading factor and monitoring a command packed in the feedback channel; and if no matching hash is found, attempting transmission at different spreading factors. The method further comprises reacting to received commands by stopping transmission, increasing spreading factor, decreasing spreading factor, or continuing transmission based on the command value.

Description

Atty. Dkt. No. 138645-0120SATELLITE COMMUNICATION SYSTEM WITH PEER-ASSISTED COVERAGE EXTENSIONCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to U.S. Application No.: 63 / 682,306, titled “SATELLITE COMMUNICATION METHODS” filed on August 12, 2024, which is hereby incorporated by reference in its entirety.BACKGROUND

[0002] Satellite communication systems have become increasingly important for providing global connectivity and enabling various applications such as navigation, remote sensing, and telecommunications. These systems typically consist of satellites orbiting the Earth and ground-based terminals that communicate with the satellites. As the demand for satellite-based services grows, there is a continuous effort to improve the coverage, capacity, and efficiency of satellite communication networks.

[0003] One challenge in satellite communications is providing reliable coverage to terminals in areas with limited line-of-sight to satellites, such as urban canyons, dense forests, or indoor environments. In these scenarios, the direct link between a terminal and a satellite may be obstructed or weakened, leading to degraded performance or loss of connectivity. Additionally, the power constraints of satellite systems and mobile terminals can limit the ability to overcome signal attenuation in challenging environments. Another aspect of satellite communication systems is the management of network resources and the optimization of communication protocols. As the number of terminals in a network increases, efficient allocation of satellite capacity and minimization of signaling overhead become increasingly important. Traditional approaches may struggle to scale effectively as network complexity grows. Furthermore, the mobility of terminals in satellite networks introduces additional complexities. As terminals move between coverage areas of different satellites or beams, handover mechanisms are required to maintain seamless connectivity. These14911-8131-1070.1Atty. Dkt. No. 138645-0120 handovers need to be managed efficiently to minimize disruptions and optimize resource utilization across the network. The integration of satellite systems with terrestrial networks is another area of focus in the field. Hybrid networks that combine satellite and terrestrial components can potentially offer improved coverage, capacity, and reliability. However, coordinating between these different network segments presents technical challenges in terms of routing, resource allocation, and quality of service management.

[0004] As satellite communication technology continues to evolve, there is an ongoing need for innovative approaches to address these challenges and improve the overall performance and capabilities of satellite networks. Advancements in areas such as signal processing, network protocols, and system architecture can contribute to enhancing the efficiency, reliability, and flexibility of satellite communication systems.SUMMARY

[0005] This summary is provided to introduce a selection of concepts in a simplified form that are further described below in the detailed description. This summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used as an aid in determining the scope of the claimed subject matter.

[0006] The present disclosure provides a satellite communication system utilizing a closed loop feedback scheme for efficient signaling. The system employs a packed control channel leveraging Forward Error Correction (FEC) with Reed Solomon algorithms to reconstruct missed data without explicit requests. Endpoints transmit protocol data units (PDUs) at different spreading factors and monitor a feedback channel for a hash matching their endpoint ID and minimum spreading factor. Upon finding a matching hash, endpoints continue transmission at the signaled spreading factor and monitor commands packed in the feedback channel, reacting accordingly to stop, increase, or decrease spreading factors.24911-8131-1070.1Atty. Dkt. No. 138645-0120

[0007] The disclosure also introduces a local -to-proxy communication method enabling multiple local endpoints to communicate uplink service data units (SDUs) to chosen proxy endpoints under contention. This method employs chip offset dimensions and frequency raster distinguishing to ensure robust communication amidst nearby proxies. Additionally, the system incorporates a subchip round trip ranging technique, utilizing coherent integration through a series of operations to optimize correlation during signal processing, enhancing timing accuracy.

[0008] Furthermore, the disclosure presents a peer assist communication scheme allowing endpoints without coverage to utilize nearby endpoints with coverage to relay messages. This approach takes advantage of an extensive link budget to address scenarios with limited coverage, such as indoor or deep basement conditions. The system also implements a power-efficient paging operation for low-latency network-initiated transactions, packing multiple paging IDs into a single frame. Lastly, an ultra-low power side channel protocol for small cells is introduced, enabling efficient indoor location tracking and asset monitoring through optimized geofencing mechanisms, significantly reducing power consumption compared to traditional GPS-based solutions.

[0009] The foregoing general description of the illustrative embodiments and the following detailed description thereof are merely exemplary aspects of the teachings of this disclosure and are not restrictive.BRIEF DESCRIPTION OF FIGURES

[0010] Non-limiting and non-exhaustive examples are described with reference to the following figures.

[0011] Figure 1 is a graphical representation of a satellite communication network architecture with various communication pathways, according to an embodiment.34911-8131-1070.1Atty. Dkt. No. 138645-0120

[0012] Figure 2 is a flowchart of a satellite communication method, according to aspects of the present disclosure.

[0013] Figure 3 is a feedback channel frame format for session ID and control signaling, according to an embodiment.

[0014] Figure 4 is a block diagram of a subchip round trip ranging system, in accordance with example embodiments.

[0015] Figure 5 is a block diagram that shows a protocol for power efficient paging which supports relatively low latency network initiated transactions, in accordance with example embodiments.

[0016] Figure 6 is a block diagram of a paging channel PDU (protocol data unit), according to aspects of the present disclosure.

[0017] Figure 7 is a graphical representation of a network communication system comparing conventional and cooperative communication paradigms, in accordance with example embodiments.

[0018] Figure 8 is a flowchart of a synchronization and communication process, according to aspects of the present disclosure.

[0019] Figure 9 is a block diagram of activity mapping presented to endpoints, according to aspects of the present disclosure.

[0020] Figure 10 is a depiction of a parallel proxy boot-strap advertisement scheme, according to an embodiment.

[0021] Figure 11 is a representation of a multi-dimensional contention scheme for L2P communication, in accordance with example embodiments.

[0022] Figure 12 is a flow diagram of a method for a repeater and small cell FRAME cycle, in accordance with example embodiments.44911-8131-1070.1Atty. Dkt. No. 138645-0120

[0023] Figure 13 is a flow diagram of a method for local -to-proxy communication contention, in accordance with example embodiments.

[0024] Figure 14 is a flow diagram of a method for local -to-proxy (L2P) communication, in accordance with example embodiments.

[0025] Figure 15 is a flow diagram of a method for proxy -to-infrastructure (P2I) and infrastructure-to-proxy (I2P) communication, in accordance with example embodiments.

[0026] Figure 16 is a flow diagram of a method for proxy -to-local (P2L) communication, in accordance with example embodiments.

[0027] Figure 17 is a protocol diagram for power-efficient paging operation between Local and Proxy Functions, according to aspects of the present disclosure.

[0028] Figure 18 is a block diagram of an implementation of a fairness algorithm , according to an embodiment.

[0029] Figure 19 depicts a block diagram of a frame structure in a communication system, according to an embodiment.

[0030] Figure 20 is a flow diagram of operations between endpoint, small cell, and cloud in a communication system, in accordance with example embodiments.

[0031] Figure 21 is a block diagram depicting generation of a channelization including frequency offset and subslot in a satellite communication system, in accordance with example embodiments.

[0032] Figure 22 is a flow diagram for a method for RTR multiple access, in accordance with example embodiments.

[0033] Figure 23 is a flow diagram for a method for paging subslot processing, in accordance with example embodiments.54911-8131-1070.1Atty. Dkt. No. 138645-0120

[0034] Figure 24 is a flow diagram for a method for round trip range response signaling and identification, in accordance with example embodiments.

[0035] Figure 25 is a block diagram showing modes of operations for firmware over the air (FOTA), according to aspects of the present disclosure.

[0036] Figure 26 is a block diagram depicting communication slots supporting FOTA in a satellite communication system, according to aspects of the present disclosure.

[0037] Figure 27 is a block diagram depicting a firmware transfer cycle protocol implemented across three subframes, according to an embodiment.DETAILED DESCRIPTION

[0038] The following description sets forth exemplary aspects of the present disclosure. It should be recognized, however, that such description is not intended as a limitation on the scope of the present disclosure. Rather, the description also encompasses combinations and modifications to those exemplary aspects described herein.

[0039] Satellite communication systems play a crucial role in enabling global connectivity and information exchange. These systems utilize orbiting satellites to relay signals between ground-based terminals, allowing communication across vast distances and in remote areas where terrestrial infrastructure may be limited or non-existent. A typical satellite communication system comprises several key components working in concert to facilitate reliable data transmission. Satellites orbiting the Earth act as relay stations, receiving signals from ground-based transmitters and retransmitting them to receivers on the ground. These satellites may operate in various orbital configurations, including geostationary, medium Earth orbit, or low Earth orbit, each offering different coverage and latency characteristics.

[0040] Ground-based terminals form another essential part of the system. These terminals may include fixed installations, mobile devices, or portable units capable of64911-8131-1070.1Atty. Dkt. No. 138645-0120 transmitting and receiving signals to and from satellites. The terminals typically incorporate antennas, transmitters, receivers, and signal processing equipment to manage the communication link with orbiting satellites. Communication protocols govern the exchange of information between ground terminals and satellites. These protocols define the structure of data packets, error correction methods, and procedures for establishing and maintaining connections. Protocols may be designed to optimize various aspects of communication, such as power efficiency, data throughput, or resistance to interference.

[0041] Satellite communication systems face unique challenges due to the long distances signals must travel and the dynamic nature of orbital mechanics. Factors such as signal propagation delay, Doppler shift, and atmospheric effects may impact communication quality and reliability. To address these challenges, advanced signal processing techniques and adaptive transmission methods may be employed. The applications of satellite communication systems span a wide range of fields, including telecommunications, broadcasting, navigation, weather monitoring, and emergency services. As technology advances, these systems continue to evolve, offering improved performance, capacity, and flexibility to meet growing global communication needs.

[0042] Figure 1 illustrates an example satellite communication system 100 including multiple components that work together to enable efficient and reliable communication across vast distances. The system 100 comprises satellites 102, endpoints 104, proxies 106, repeaters 108, and small cells 110, each playing a specific role in the overall network architecture. The satellite 102 is positioned in orbit to facilitate communication between ground-based components. The satellite 102 supports millions of endpoints through direct communication paths. In some implementations, an endpoint transmits a set of protocol data units (PDUs) at different spreading factors with a session request bit. The satellite 102 receives a subset of the PDUs based on link conditions.74911-8131-1070.1Atty. Dkt. No. 138645-0120

[0043] Endpoints represent various devices capable of communicating with the satellite 102 or other network components. Endpoints may include local (L) endpoints 104 and proxies (P) 106. Endpoints 104 may refer to devices performing normal application operations. Endpoints 104 may include mobile devices, fixed terminals, or other communication equipment designed to operate within the satellite network. Proxies 106 may refer to an endpoint’s role in assisting local operations of other local endpoints 104. That is, proxies 106 may serve as intermediary nodes in the communication network. In some implementations, proxies 106 are endpoints with favorable signal conditions that relay messages for other endpoints with weaker connections. This proxy functionality extends the effective range and coverage of the satellite communication system 100. In other embodiments, that endpoints may additionally hop through multiple other endpoints before transmission to the network.

[0044] Repeaters 108 are deployed to support typical communication with multiple endpoints. These repeaters 108 amplify and retransmit signals, helping to overcome signal degradation over long distances or in challenging environments. The repeaters 108 may be powered endpoints that are always operational and dedicated to being full-time “proxy endpoints.” Small cells 110 are internet-connected endpoints that typically support a plurality of endpoints. These small cells 110 provide localized coverage and help offload traffic from the satellite, improving overall network efficiency. A small cell 110 may always be operational and may be connected to the network via integrated Wi-Fi or Ethernet. As such, a small cell 110 may be similar to a repeater 108 having a Wi-Fi or Ethernet connection.

[0045] The system 100 incorporates flexible routing of communications through multiple possible paths. Endpoints 104 establish optimal communication links based on their location and conditions, choosing between direct satellite connection, terrestrial connection, repeater connection, or small cell connection based on which communication link is strongest. The network utilizes a terrestrial component 112 that provides an alternative communication path through broadband connections. This terrestrial infrastructure complements the satellite-84911-8131-1070.1Atty. Dkt. No. 138645-0120 based communication, offering redundancy and increased capacity in certain areas. In an embodiment in which the endpoint 104 considers only power consumption optimization, the endpoint 104 may link to the component (e.g., satellite 102, repeater 108, small cell 110, or terrestrial component 112) having the strongest link. In embodiments in which a weight is considered for the capacity of the satellite 102, the endpoint 104 may prefer one of the terrestrial access point 112 or the small cell 110 for linking, even if the link is relatively weaker than a satellite or repeater link.

[0046] In various embodiments, if a proxy reverse link channel is considered, the selection of which component to communicate with may depend on a sum (or any other mathematical operation) of the power of the endpoint 104 and an estimated power of the proxy 106, based on a link quality between the proxy 106 and the infrastructure. Based on the sum being less than the power consumption by the endpoint 104 of any of the satellite 102, the repeater 108, the small cell 110, or the terrestrial access point 112, the proxy 106 may be selected.

[0047] The communication paths between components are bidirectional, allowing for both uplink and downlink transmissions. The system 100 dynamically adjusts these paths based on network conditions, endpoint capabilities, and available resources to optimize overall performance.

[0048] The system 100 extends network coverage and improves efficiency in challenging environments using proxy endpoints, repeaters, and small cells. Proxy endpoints serve as intermediary nodes in the communication network. These are endpoints with favorable signal conditions that relay messages for other endpoints with weaker connections. This functionality extends the effective range and coverage of the satellite communication system, particularly in areas where direct satellite communication is difficult or impossible.94911-8131-1070.1Atty. Dkt. No. 138645-0120

[0049] Figure 2 illustrates a method 200 for bit-efficient communication between endpoints and satellites using an optimized closed loop feedback scheme. In this optimized closed loop feedback scheme, commands are efficiently packed into downlink signaling to support uplink communication, particularly in a paired frequency spectrum where channel reciprocity may not be utilized. The scheme or method 200 described with respect to Figure 2 layers on top of a fragmentation scheme that leverages Forward Error Correction (FEC) (e.g., erasure-based reconstruction code) with Reed Solomon algorithms, which enables a network to reconstruct missing data without explicit requests for specific missed pieces. It should be understood that the present disclosure is not limited to the use of Reed Solomon algorithms and any other suitable algorithm or algorithms may be used.

[0050] During a session, an endpoint transmits a set of protocol data units (PDUs) at different spreading factors. For example, at step 201, multiple PDUs are transmitted in the same frame at multiple different spreading factors. Each PDU includes a session request bit 202 in its header. In some examples, the PDUs are unique codeword segments of the Reed Solomon FEC. A satellite receives a subset of the transmitted PDUs, referred to as received PDUs, at step 203, based on link conditions. Upon receiving the PDUs, the satellite identifies a minimum spreading factor PDU that was received. The satellite constructs a hash H based on the endpoint ID and the identified minimum spreading factor at step 204. The hash may be, for example, a puncture hash. In an example, the hash output is 16 bits. However, in various other examples, the hash output may be more or less than 16 bits. This endpoint session ID hash 205 is packed with other data in a feedback channel that all active endpoints monitor.

[0051] The endpoint receives the common feedback channel and searches for any hash H that is consistent with the endpoint’s ID across all spreading factors that the endpoint most recently transmitted. If such a hash is found, the endpoint continues transmitting PDUs in the codeword at the signaled spreading factor, at step 206. The bit position where the valid hash104911-8131-1070.1Atty. Dkt. No. 138645-0120H is found may correspond to a Session ID assignment. The Session ID assignment may be a value between Session 0 and Session 81. If no hash is found, the endpoint attempts another set of spreading factors at step 207, possibly increasing to the maximum spreading factor possible.

[0052] Once the hash is received, the endpoint monitors the 2-bit command C packed in incrementing session ordering corresponding to the endpoint’s Session ID at step 208. The endpoint reacts to the received command based on the content of the received command. For example, if the command is “00,” indicating a “stop” command, the endpoint stops transmission. A stop command occurs when the network determines, at step 209, that a sufficient number of PDUs have been received such that the codeword can be algorithmically reconstructed. Upon receipt of the “stop” command by the endpoint, the endpoint may consider the SDU successfully received. The endpoint may then move on to transmitting the next SDU at step 210, or go to sleep at step 211. In another embodiment, the destination of the 2-bit command may be on a mutually (e.g., between endpoint and network) understood scheduling.

[0053] In some examples, the command received at step 212 by the endpoint is “01,” indicating an “increase spreading factor” command. An increase spreading factor command may be received or occur when the satellite fails to receive any PDUs in a given frame, or if the margin on the received PDUs is below a certain threshold. Upon receipt of an increase spreading factor command, at step 213, the endpoint adjusts the spreading factor. In an example embodiment, the endpoint may increase the spreading factor by a factor of 2 (e.g., allowing 3 dB more link budget).

[0054] In some embodiments, the command received at step 214 by the endpoint is “10,” indicating a “decrease spreading factor’ command. A decrease spreading factor command may occur when the satellite receives PDUs above a certain threshold. As such, at step 215, the endpoint may decrease the spreading factor. The spreading factor may be decreased, for114911-8131-1070.1Atty. Dkt. No. 138645-0120 example, by a factor of 2 (e.g., causing 3 dB less link budget). In some embodiments, the command received at step 216 is “11,” indicating a “no command” command. A “no command” command indicates the satellite has used those bits for hash signaling and there is no command for the endpoint. In such an embodiment, at step 217, the endpoint may continue to transmit at the same spreading factor. An optimization is shown. In some examples, a hash value exists that is an immediate stop 218. This immediate stop may be issued when enough PDUs have been received by the network in the initial set of PDUs transmitted. In such a scenario, the endpoint can cease transmitting the current SDU and may not need a Session ID. Another optimization may be a hash that signals “Rewind” based upon the previous SDU not being successfully received. This may occur due to, for example, a hash collision. In this case, the endpoint reverts to the previous SDU and continues transmitting PDUs where it left off. In an alternative embodiment, the acknowledgement of a single PDU message may be acknowledged by a 2 -bit command C based on the mutually known Session ID.

[0055] Figure 3 illustrates a structure of an example feedback channel 300 including a frame format that supports both hash and commands. The feedback channel 300 includes a header field 301 that signals valid session command fields. A session hash 302 occupies multiple session slots within the feedback channel. The hash field 303 contains the endpoint session ID hash. A hash sequence 304 represents multiple hashes within the feedback channel. A session wrap field 305 indicates where hash values corresponding to higher session numbers wrap around to overwrite initial sessions.

[0056] The feedback channel is monitored by multiple active endpoints. A bit position where the matching hash is found corresponds to a session ID assignment. The feedback channel is sent at a selectable spreading factor other than the maximum allowed to conserve downlink capacity. In some implementations, the feedback channel is repeated at multiple spreading factors to maximize the number of commands and hashes that can be conveyed. The satellite communication system accounts for endpoints transitioning between beams. An124911-8131-1070.1Atty. Dkt. No. 138645-0120 endpoint transitioning between two beams while a session ID is assigned relinquishes the session ID if there is a collision with another endpoint already active in the new beam.

[0057] The hash constructed by the satellite is based on the minimum spreading factor PDU that was received. In some implementations, the hash includes additional commands, such as an immediate stop command or a rewind command to revert to a previous SDU. A hash in a given location, for example Session ID 3 302, overwrites Session 4 through Session 11 303. In this example, the hash corresponds to a Session ID of 3, consistent with the hash of a particular endpoint recently (e.g., within a certain time period) received with the minimum spreading factor that the satellite received from that endpoint. Based on the signaling of the bit pattern “11” in the command field 302 (specifically indicating Session 3 in Figure 3), endpoints corresponding to Sessions 4 through 11 understand that there is no issued command. A hash corresponding to Session 74 or greater (indicated at 305 in Figure 3) wraps to overwrite the first sessions. For the endpoint to resolve ambiguities associated with this, a header field 301 signals a valid session command field. A valid session command field may be a field that is not a portion of a hash.

[0058] Various refinements can be incorporated into this algorithm. For example, the feedback channel may send out a selectable spreading factor other than the maximum allowed spreading factor. This may conserve downlink capacity. In addition, the feedback command channel may repeat at multiple spreading factors. This may maximize the number of commands and hashes. Further, an endpoint may transition between two beams while a session ID is assigned. This may cause or lead to a session ID collision with another endpoint already active in the new beam. A session ID assignment hash matching the endpoint ID already active in the new beam may cause the newly entering endpoint to relinquish the session ID.Subchip Round Trip Ranging134911-8131-1070.1Atty. Dkt. No. 138645-0120

[0059] Figure 4 illustrates the components and processes of a subchip round trip ranging system 400. Specifically, the system 400 incorporates a sub-sampled pulse-shape filter operating at, for example, a cx2 rate, to enable precise timing measurements in satellite communications. The filter processes pulse shape values delayed by, for example, j / 512 of a chip. The pulse-shape filter accounts for both transmit and receive filtering effects, with an increased bandwidth-time (BT) factor by, for example, sqrt(2), compared to individual filters. This combined filtering approach optimizes the signal characteristics for subsequent processing.

[0060] The system performs a matrix multiplication operation between two primary components: a pulse shape filter matrix Pij and a sync symbol correlation vector Vj operating at, for example, a 1 ps rate. The multiplication of these components produces a sub-chip hypotheses metric vector Mi. The system determines timing by finding the maximum value in the metric vector Mi, which corresponds to the most likely sub-chip hypothesis, providing a high-resolution estimate of the signal timing. Specifically, the i of a maximum value in the Mi vector may be the likeliest sub-chip hypothesis.

[0061] The core algorithm employs sufficient coherent integration to ensure the various subchip correlations have positive signal to noise ratios (SNRs). A positive SNR as discussed herein refers to an SNR ratio less than the noise squared terms which may dominate in a conversion to non-coherence. The process follows the following steps: first, 4-8 symbols of coherent combine as an inner loop. Second, an I2+Q2operation is performed. A non-coherent accumulation is performed across the entire frame or sync symbol. In alternative embodiments, all of the un-wiped symbols may be input into a Fourier Transform (FT) or Fast Fourier Transform (FFT) to learn the Sync Symbol Correlation Vector with a potential outer loop that tests various quantized rates of time tracking drift.

[0062] The system generates sums at chip x 2 spacing around the peak of the pulse shape. In some embodiments, the system may generate sums at chip x 4 spacing. As shown in Figure144911-8131-1070.1Atty. Dkt. No. 138645-01204, Vj may be a vector of these magnitude sums at cx2 spacing. Pij may be a matrix in which each row is the expected magnitude of voltage for a given subchip hypothesis j over the cx2 spacings i. The determination involves forming the product P x V and finding j which maximizes this quantity, corresponding to the subchip timing nominally in units of cx512. For example, assuming + / - cx4 (where J is 256) and 4 cx2 correlations (where I is 4), this corresponds to a matrix of 256x4 = 1024 coefficients.

[0063] For low Earth orbit (LEO) satellite implementations, the system addresses the challenge of measuring a moving target of subchip phase. The compensation for time tracking drift uses the formula T = mean(Fj - j*TTj - Cj), where TTj is the relative time tracking drift in units of sub-chip granularity derived from the measured frequency offset, and Cj is the total time tracking correction applied by the modem. The result is the sub-chip phase at the instant of the beginning of signal transmission. For uplink processing, the system subdivides segments on super symbol boundaries, for example with J=384 symbols / 8 super symbols = 48, requiring 48 matrix multiplications per frame. To reduce computational requirements, super symbols may be non-coherently combined to form segments larger than super symbols. As an example, all symbols in the frame may be combined via a Fourier Transform (FT) or Fast Fourier Transform (FFT). For downlink processing, a dynamic super symbol length may be based on a signal to noise ratio (SNR). The number of symbols per super symbol may be reduced to reduce spectral smearing.Power Efficient Paging Operation

[0064] Figure 5 illustrates the key components and processes of a power-efficient paging operation system to optimize communication while minimizing power consumption. Specifically, Figure 5 illustrates a method 500 for communication between an endpoint and a satellite or tower for power-efficient paging. The method 500 may support transactions initiated in a relatively low-latency network. The system assigns paging slots to endpoints based on specific criteria. At step 502, an endpoint initiates the method 500 by sending a page154911-8131-1070.1Atty. Dkt. No. 138645-0120 registration request to the satellite / tower infrastructure. The endpoint may request a paging slot at a particular paging rate. In response, at step 504, the satellite / tower issues a page registration grant. In the page registration grant, the satellite assigns a paging slot phase and frequency for the endpoint to monitor. The satellite further assigns a paging identifier for the endpoint to monitor. The paging identifier may be, for example, a 16 bit paging identifier. The assignment method by the satellite reduces unnecessary wake-up cycles for the endpoint, contributing to improved overall power efficiency. The system incorporates multiple monitoring paths to balance power efficiency and communication reliability.

[0065] At step 506, a slotted paging monitor operates at low power consumption. Specifically, the endpoint attempts demodulation of one or more low spreading factor paging channels. Demodulation may be performed at an assigned (e.g., predetermined) phase. Additionally, a slotted paging monitor with deep coverage capabilities activates when improved link conditions (e.g., when one or more metrics associated with a communication link are above a threshold level of performance) are detected. These monitoring paths allow the system to adapt to varying signal conditions while minimizing power usage.

[0066] Service data unit (SDU) transactions form a critical part of the communication process. The endpoint initiates an SDU transaction when data or commands are present. This transaction corresponds with a matching SDU transaction at the satellite / tower. The satellite / tower determines the paging slot timing based on the known location of the endpoint within a particular beam and assigns a serving satellite to monitor the page if data is present at the time of the paging slot. This coordinated approach minimizes unnecessary transmissions and receptions.

[0067] To further optimize power consumption, the system employs, at step 508, capacity efficient paging channels operating at spreading factor of, for example, 128. Specifically, based on a known location of the endpoint, the serving satellite sends a page if data is determined to be present and the current time is the time of the paging slot for the164911-8131-1070.1Atty. Dkt. No. 138645-0120 particular endpoint. At step 510, the system employs coverage optimized paging channels operating at a spreading factor of, for example, 1024. Step 510 may occur responsive to a failed endpoint response to a page at step 508. The capacity efficient and coverage optimized paging channels work in tandem to provide efficient power consumption while maintaining reliable communication between the endpoint and satellite / tower infrastructure. When the satellite employs coverage optimized paging channels at 510 and data and / or a command is present for the endpoint, the method 500 continues to step 516 where an SDU transaction occurs.

[0068] At step 512, the slotted paging monitor operates at deep coverage. Specifically, at step 506, while the slotted paging monitor operates at a low power consumption, a paging identifier (generated at step 504) may be matched. Responsive to such a match, at step 514, a SDU transaction occurs. At step 506, when there is no match between a paging identifier, the method 500 continues to step 512. That is, step 512 occurs responsive to m number of consecutive paging synchronization demodulations occurring with no success (e.g., no paging identifier matches). While the slotted paging monitor is operating with deep coverage, responsive to a match with a paging identifier, the method 500 continues to step 514 where a transaction occurs.

[0069] The system maintains communication reliability through a bidirectional interaction between the endpoint and satellite / tower. As shown, successful page acknowledgments and failed endpoint responses trigger appropriate system responses to ensure consistent paging operation. This adaptive approach allows the system to balance power efficiency with communication integrity.

[0070] Figure 6 illustrates a single frame 600 containing multiple paging IDs 602 for a paging channel PDU. For example, a paging channel PDU (as described above with respect to Figure 5) may pack multiple paging IDs into a single frame. The number of paging IDs packed into a frame may be, for example, 8 paging IDs.174911-8131-1070.1Atty. Dkt. No. 138645-0120Peer Assist

[0071] A peer assist framework is discussed herein. A peer assist concept allows for an endpoint without satellite or terrestrial coverage (referred to herein as a challenged endpoint) and within a link budget of a peer endpoint with satellite / terrestrial coverage (referred to herein as a proxy endpoint) to relay a reverse link message from the challenged endpoint to the proxy endpoint. The message may subsequently be transmitted from the proxy endpoint to the satellite / terrestrial infrastructure. The forward link may also be supported in that the satellite / terrestrial infrastructure may convey that message to proxy endpoint. The proxy endpoint may then relay the message back to the challenged endpoint. Additionally, the peer assist process may be utilized in an embodiment in which the challenged endpoint has a weak link to the satellite / terrestrial infrastructure and chooses to bounce the signal to a proxy endpoint that has a stronger link. This may conserve battery life and potentially minimize downlink capacity.

[0072] As an example, one or more endpoints may be located within a container within a cargo hold. Having the ability to leverage more favorably located endpoints to hop through may solve problems associated with a lack of coverage.

[0073] The peer assist approach may include an ability of the endpoint to perform various types of data transmission. For example, peer assist may allow an endpoint to perform Sync transmission (TX), in which signals are transmitted in a coordinated manner such that there is an alignment in the timing between a transmitter and receiver. Peer assist may also allow an endpoint to perform a Broadcast Channel (BCH) transmission. In a BCH transmission, a single source may transmit data to multiple receivers simultaneously over a shared medium. Further, peer assist may allow for DL Unicast transmission. DI Unicast transmission includes transmitting data over a DL channel to a single receiver (RX), as opposed to multiple receivers. Further, the proxy may be a pass-through, thereby eliminating a need for additional security. In addition, the satellite and / or gateway may remain unchanged.184911-8131-1070.1Atty. Dkt. No. 138645-0120

[0074] In some implementations, a peer-to-peer (P2P) scheme may be optimized. For example, the P2P scheme may be optimized in a low density area where a single proxy device (e.g., proxy endpoint) is located within a range of a challenged device or endpoint. As another example, the P2P scheme may be optimized in a high density area where there exists a number of endpoints at or above a threshold value.

[0075] The interaction at that point between a proxy endpoint and a challenged endpoint may operate at relatively high capacity by employing various fundamental communication layer elements. For example, a transmit power control of the challenged endpoint to the proxy endpoint may be implemented. Further, an erasure-based FEC and / or dynamic data rate adaptation may be implemented.

[0076] Figure 7 illustrates a system 700 for a peer assist communication architecture. The system 700 demonstrates the benefits of a peer-assisted system 710 compared to a conventional direct communication (e.g., cellular) system 702. Specifically, Figure 7 shows a transmission rate for advantaged endpoints and an indoor transmission rate for challenged endpoints. In a conventional approach, illustrated by the cellular approach system 702, advantaged endpoints transmit at an advantaged transmission energy, while indoor endpoints require a significantly higher indoor transmission energy (e.g., 533 Joules versus 8 Joules for approach described by the peer-assisted system 710) to close the link. The peer-assisted approach shown as the system 710 illustrates a proxy transmission energy, which enables more efficient data transfer for indoor devices. Transmission bits represent the data flow through the network.

[0077] As shown by system 702, endpoints communicate directly to a cell tower 704. A data rate is optimized to link conditions where the data rate is the maximum rate that is sufficient to close the link directly to the infrastructure. As shown, an advantaged endpoint 706 may have favorable endpoint placement. An advantaged endpoint 706 may beneficially have an increased data rate relative to a challenged endpoint 708 having a challenged endpoint194911-8131-1070.1Atty. Dkt. No. 138645-0120 placement. A favorable endpoint placement may be one in which the advantaged endpoint 706 is positioned such that the cell tower is visible to the advantaged endpoint 706 and can more easily connect without blockage. A challenged endpoint placement may be one in which the challenged endpoint 708 is positioned such that the cell tower is not visible to the challenged endpoint 708 and is blocked (e.g., by a physical structure). As such, a favorable indoor endpoint placement may be near a window or external wall of a building. A challenged endpoint placement may be deep inside of a building (e.g., in the center of a building, blocked by walls, etc.).

[0078] As stated, an advantaged endpoint 706 may have a higher data rate to close the link to the cell tower relative to a challenged endpoint 708 (e.g., 2 kbps versus 30 bps). As a result, there exists a difference in an amount of transmit energy required to send data from each of the advantaged and challenged endpoints. For example, for the advantaged endpoints 706, 8 Joules of energy may be used to send a kilobyte of data, while for challenged endpoints 708, a higher amount of energy (e.g., 533 Joules versus 8 J) is required to send a kilobyte of data. In various embodiments, a battery may have a set amount of energy (e.g., Joules) that can be delivered. Accordingly, in the system 702, there may exist a large difference in rate of battery drain. The mean of energy per kilobyte between the advantaged endpoints 706 and the challenged endpoints 708 may be, for example, 270 Joules, and the standard deviation may be, for example, 371 Joules.

[0079] The peer-assisted communication system 710 optimizes power consumption and extends battery life for endpoints in challenging environments. As shown, the peer-assisted communication system 710 leverages proxies with better signal conditions. That is, the advantaged endpoints 706 “hop” data from the challenged endpoints 708 through to a satellite 712. For example, the satellite 712 may transmit and receive data from an advantaged endpoint 706. The advantaged endpoint 706 may, in turn, transmit and receive data from a challenged endpoint 708. In this manner, the challenged endpoints 708 indirectly204911-8131-1070.1Atty. Dkt. No. 138645-0120 communicate with the satellite 712 via the advantaged endpoints 706. As a result, the system 710 reduces the transmission power required for endpoints with poor direct satellite visibility (e.g., challenged endpoints 708). This power reduction may result in a longer battery life and improved overall system efficiency. The assisting endpoint (e.g., advantaged endpoint 706) may require or utilize more transmit power (e.g., relative to an amount of transmit power required by the advantaged endpoint 706 in the system 702) to hop the data from the challenged endpoint 708 to the satellite 712. However, the reduction of transmit power required for the challenged endpoints 708 may be larger than the increased transmit power required by the advantaged endpoints 706. The result may be a decrease in both the mean (and standard deviation.

[0080] Figure 8 illustrates the flow of a peer-to-peer process 800, detailing how endpoints associate with proxies and communicate through the network. The process begins at step 802 with an HOUR advertisement. As used herein, “HOUR” refers to a rate at which endpoints can discover new peers (e.g., endpoint-based devices that extends coverage, such as a proxy endpoint, a repeater, or a small cell) when they are shifted to a new environment. An HOUR may be, for example, a period of 3600 seconds. Demodulation syncs may occur on the HOUR (e.g., every 3600 seconds) and a MINUTE may be coded. As used herein, a “MINUTE” may refer to an eligible beat in which proxy operation can begin. A “beat” as referred to herein may be a peer frame rate (proxy, repeater, small cell). A beat may be or include a 1 second down beat (forward peerlink) and / or 1 second up beat (reverse peerlink).

[0081] Forward communication may be communication originating from a backend and being transmitted to an access point, then to a proxy, then to an endpoint. Reverse communication may be communication originating from an endpoint and traveling to a proxy, to an access point, and finally to the backend. Further, a “second” as referred to herein may refer to a fundamental tick rate. Even parities may be forward “downlink” and odd parities may be reverse “uplink.” A “second” may have a 1 second rate synchronized with a day,214911-8131-1070.1Atty. Dkt. No. 138645-0120HOUR, MINUTE, frame, etc. A frame may be an access point (e.g., satellite or terrestrial infrastructure) frame rate. A frame may be generated or transmitted, for example, every six seconds. A frame may be synchronized to a day, HOUR, and / or MINUTE.

[0082] Referring again to step 802, a normalized link quality (NLQ) may be learned. An NLQ may be a normalized link quality from a proxy to an infrastructure. The NLQ may be a measure of a link quality that is normalized in some manner (e.g., by an activity factor). Step 802 may allow endpoints to learn up to, for example, 480 potential peers for that HOUR. For example, 60 minutes in the HOUR and 8 NLQs may equal 60 x 8 = 480 peers. The system incorporates a contention mechanism for endpoints to select the best proxy for communication.

[0083] Following the HOUR advertisement, the system checks for a pending uplink service data unit (UL SDU). If an UL SDU is pending, the process moves to step 804. At step 804, a MINUTE-sync step is performed. In this step 804, responsive to a determination that a local endpoint has a pending UL PDU, sync symbols are demodulated at the first known MINUTE timing signaled in the HOUR advertisement across the NLQ slots. Local endpoints without timing may attempt to recover timing via periodic sync processing. As such, the step 802 may repeat.

[0084] The process then enters a local -to-proxy (L2P) contention phase in step 806. During this phase, a local endpoint monitors all proxies with superior NLQ by demodulating across the NLQ syncs (e.g., 10 NLQ syncs). The system selects the best proxy based on a metric. As used herein, a “metric” is a function of signal strength (RSSI), proxy normalized link quality (NLQ), and loading (activity factor). The local endpoint then contends with the proxy that has the best metric. When the local endpoint determined that its offset is signaled in a sync, contention is granted.224911-8131-1070.1Atty. Dkt. No. 138645-0120

[0085] Once contention is resolved or granted, the process moves to a L2P communication step 808. In step 808, the local endpoint employs forward error correction (FEC) to communicate the UL PDU via FEC until an acknowledgment (ACK) is signaled in the sync symbol. During step 808, the local endpoint packs in multiple PDUs using FEC on downlink unicast PHY. The local endpoint may pack in the multiple PDUs in a serial manner. The PDUs may be packed until a sync acknowledgement is received.

[0086] After L2P communication, the system proceeds to a proxy-to- infrastructure / infrastructure-to-proxy (P2I / I2P) communication phase in step 810. During step 810, the proxy rolls through the identity of all associated local endpoints serially and performs Uplink / Downlink (UL / DL) communication to the infrastructure.

[0087] The next phase is a proxy -to-local (P2L) communication in step 812. Step 812 may occur responsive to all UL PDUs being processed and any DL PDUs being received by the proxy. In step 812, the proxy transmits a sync at a specific offset to signal to set up a channel for the first local endpoint to service at a current hopped frequency at the first MINUTE past P2EI2P. The proxy continues to transmit the sync until all local endpoints are serviced, or a timeout occurs. In serial fashion, the proxy employs FEC to communicate UL ACK and DL SDU until the local endpoint sends an ACK message.

[0088] The process 800 enters a sleep state in step 814. Before entering sleep, the system ensures that all UL PDUs are processed and any DL PDUs received by the proxy are handled. The system remains in the sleep state until the next MINUTE, at which point the cycle begins again. Throughout the process, the system employs specific timing mechanisms and frequency hopping patterns to optimize communication efficiency and minimize power consumption. The method allows for dynamic adaptation to changing network conditions and endpoint requirements, ensuring robust and reliable satellite communication.234911-8131-1070.1Atty. Dkt. No. 138645-0120

[0089] Figures 9 and 10 illustrate components and processes of a proxy boot-strap advertisement scheme. The scheme presents a global mapping of activity to all endpoints each HOUR through a known channel. Each forward peerlink downbeat, starting from the HOUR, organizes a set of Syncs in the frame structure. First, as shown in the scheme 900, each proxy transmits a timing recovery 902. The timing recovery may consist of a, for example) 128 symbol TX Sync Symbol Sequence (e.g., a subset window of 32 symbols of those symbols received) on a known channel with no tuning or frequency offset. This occurs at the beginning of the first beat of the hour, enabling endpoints to establish absolute HOUR timing and perform TCXO checks.

[0090] Following the timing recovery, a MINUTE encoding 904 occurs using, for example, a 40 symbol TX Sync Symbol (RX 32) modulated by a frequency offset that signifies the MINUTE of activity. Based on the NLQ, a specific beat is selected for a given proxy to transmit the sync energy. The Sync Slots are packed two into a beat in a specific order. The order may be, for example, Small Cell, Repeater, and Link Quality 0 through Normalized Link Quality 7, representing the strength of link to infrastructure from best to worst, as shown in Figure 9. Multiple (e.g., 10) sync slots may fit into a one second forward beat.

[0091] Referring to Figure 10, a parallel proxy boot-strap advertisement scheme 1000 is shown. The timing recovery is sized to allow endpoints to decimate operation by a factor of, for example, 4 (128 TX symbols / 32 RX symbols processed). Because the Sync symbols are unmodulated, a narrow frequency range search is sufficient and can be searched while consuming an amount of power below a particular level. As such, an average power consumption during timing recovery that is a fraction of the power consumed (e.g., approximately % when the factor by which to decimate operation is 4). Each Sync at the appropriate NLQ slot is tuned to a frequency, for example, at 5 kHz rasters, that signals the minute offset. For example, 150 kHz corresponds to 0 minutes after the hour, while +150 kHz244911-8131-1070.1Atty. Dkt. No. 138645-0120 corresponds to 59 minutes after the hour. This frequency encoding allows endpoints to learn multiple proxies per given HOUR / NLQ slot.

[0092] The local endpoint may learn multiple proxies by first performing detection on the first Sync and identifying the strongest chip timing for each Doppler hypothesis. No detection finger processing is required for this process, although normal detect finger processing is useful for timing recovery to realign Symbol timing. When the winning chip timing for a given Doppler hypothesis exceeds a threshold, the system identifies it as a valid signal. The Doppler hypothesis is mapped to the nearest 5 kHz raster, for example, which signals the active MINUTE of that proxy's service.

[0093] The system takes the delta error from the 5 kHz rasters for each value and applies a correction to the endpoint's timing. This correction may be implemented as feedback in an HR style with, for example, a 1% alpha value. The approach also provides endpoints entering a new unfamiliar environment with decision-making capability regarding available communication options at a reasonable (e.g., hourly) latency rate. The system leverages signal processing capabilities to enable efficient discovery of nearby proxy nodes. The frequency allocation strategy optimizes the use of available spectrum. Each proxy operates on a distinct frequency range, allowing for simultaneous transmission without interference. The frequency separation enables receiving nodes to distinguish between different proxy advertisements easily.

[0094] By implementing this parallel approach, the proxy boot-strap advertisement scheme 1000 allows for simultaneous processing of multiple proxy connections while maintaining timing synchronization across the network. This parallelism reduces the time required for network discovery and initialization, leading to faster and more efficient establishment of communication links within the satellite network. The proxy boot-strap advertisement scheme directly maps frequencies to specific minute advice timing, creating a time-based transmission schedule that allows receiving nodes to predict and efficiently scan254911-8131-1070.1Atty. Dkt. No. 138645-0120 for proxy advertisements. The small cell ultra-low power side channel protocol for location determination utilizes a specialized frame structure and operational flow to enable efficient communication between endpoints, small cells, and the cloud system. This protocol optimizes power consumption while maintaining accurate location tracking capabilities.Local-to-Proxy (L2P) and Proxy-to-Local (P2L)

[0095] Figure 11 illustrates a multi-dimensional contention scheme 1100 for local-to- proxy (L2P) communication, which optimizes network efficiency and collision resolution. The diagram depicts three distinct beats labeled “Beat 1 UL,” “Beat 1 DL,” and “Beat 2 DL,” arranged horizontally to show the temporal progression of the contention process. The BEAT structure in the scheme 1100 occurs every second and differs compared to typical access point frame structures. There are two types of beats: downbeats for forward peerlink communication (e.g., proxy infrastructure transmits, challenged endpoint receives) and upbeats for reverse peerlink communication (e.g., challenged endpoint transmits, proxy infrastructure receives).

[0096] In the L2P contention process, local endpoints compete for access to proxy endpoints based on a metric, which is a function of signal strength (RSSI) and NLQ. This ensures that endpoints connect to the most suitable proxy, balancing link quality and network load. The contention mechanism allows local endpoints to demodulate, for example, 10 SYNC SLOTS in a given second, choosing the strongest existing metric for sending its uplink PDU. The frequency hopping pattern, determined by a hash of the proxy’s current NLQ and seconds in the hour, adds another dimension to the contention scheme. All proxies with the same NLQ hop together in the same pattern (e.g., including additional proxies coming online at future MINUTE ticks), while local endpoints use a time offset dimension to determine proxy identity when multiple proxies have the same NLQ. This approach allows for efficient use of the available spectrum and enables local endpoints to dynamically change which proxies they contend with based on the metric. Lock and unlock may be applied at different rates. Further, in various embodiments, local endpoints can dynamically change which proxies they are contending with based on a metric, which includes, in some examples,264911-8131-1070.1Atty. Dkt. No. 138645-0120 handing over to a different proxy in the same NLQ (e.g., communicating with the proxy that is on a different chip phase).

[0097] If a local endpoint loses contention, it hops to the next frequency in the hopping sequence. At scale, multiple local endpoints can contend with different local endpoints at the same NLQ, supporting a first number of proxy subjects and a second, larger number of local contentions simultaneously. That is, the prior proxy that the local endpoint was contending with may be locked and unreachable, allowing another proxy to contend for. The local endpoint may only contend on proxies being at an NQL less that its current NQL. The scheme 1100 also illustrates the transition between different communication phases. When a proxy is not hit with additional contentions after a TIMEOUT value, it transitions to the proxy-to- infrastructure (P2I) and infrastructure-to-proxy (I2P) phase. Upon completion of this phase, at the following MINUTE, a Sync is transmitted at the current frequency, triggering local endpoints to attempt contention on that proxy. Local endpoints serially win contention and receive their UL SDU and ACK DL SDU if present.

[0098] This multi-dimensional approach, combining time (BEAT structure), frequency (hopping patterns), and signal quality (METRIC) dimensions, creates a robust and efficient contention mechanism. It allows the system to optimize network resources, minimize collisions, and adapt to varying network conditions, ultimately enhancing the overall performance and reliability of the satellite communication system.Repeater and Small Cell FRAME Cycle

[0099] Referring now to Figure 12, a flow diagram illustrating a process 1200 for a repeater and small cell FRAME cycle, according to an example embodiment. For a repeater and small cell frame cycle, based on the parity of the frame, the repeater operates in either an L2P / P2L mode or a P2I / I2P mode. At step 1202, a repeater gathers and / or disburses SDUs to local endpoints followed by, at step 1204, UL / DL transactions to the satellite / tower. This occurs at a frame rate. As a result, paging indicators match the parity of the repeater. This is taken into account for the paging channel design described herein.274911-8131-1070.1Atty. Dkt. No. 138645-0120

[0100] For the FRAME parity that is P2L / L2P, the first DL beat relays, at step 1206, the paging channel. This step may be optional (e.g., may occur only when FRAME parity is P2L / L2P). The second DL beat is L2P and the third DL beat is P2L. Local endpoints experience at, for example, a % decimation rate relative to peer-peer interactions for L2P and P2L.

[0101] For Small Cell, the pattern may not include a P2I / I2P phase. In various examples, % bandwidth is a paging channel at a maximum SF (both parities), with the remaining bandwidth split evenly (e.g., 1 / 3 each) between L2P and P2L.Local-to-Proxy Communication (L2P) Contention

[0102] Referring now to Figure 13, a method 1300 illustrating local -to-proxy communication contention. In a local -to-proxy (L2P) communication contention, at step 1302, multiple local endpoints communicate their UL SDU to a chosen proxy endpoint. This may occur, for each endpoint, in the presence of multiple nearby proxies. Each proxy may also be in communication with its own set of local endpoints that are communicating their UL SDUs to the proxy. In various embodiments, a chip offset dimension distinguishes local endpoints in a given proxy subnet. The frequency raster distinguishes between proxy subnets. The process by which this distinction is made is described herein.

[0103] At step 1304, a proxy may first send a Sync in a first D phase at a sync phase corresponding to the proxy’s NQL. Each slot phase may be, for example, around 1 second. At step 1306, all local endpoints may contend for the channel in the U phase by sending a Sync. The Sync may be “normally transmit power controlled” at a frequency FL “Normally transmit power controlled” as used herein may indicate that the Sync is only reduced to the level of the noise floor upon reception by the proxy and no further. The Sync may also be controlled at a chip timing of 100R. In various embodiments, R may be between 0 and 100. Alternatively or additionally, endpoints put a frequency offset raster (FOR) to additionally reduce ambiguity. The frequency modulation may be echoed in the CONTENTION RESPONSE.284911-8131-1070.1Atty. Dkt. No. 138645-0120

[0104] At step 1308, in a subsequent D phase, the proxy transmits a Sync based on the chip timing of the strongest signal received. The limiter of distance of the delay at an endpoint proxy may be less than, for example, 90 chips. In various embodiments, 90 chips corresponds to 17 miles. Thus, a distance limit may be less than or equal to 17 miles. Further, a contention collision may yield a false ACK to the weaker endpoint signal. However, if the weaker signal takes longer to transmit its PDUs, the ACK may be likely to be observed before the systematics are transmitted. As such, the weaker local endpoint may disregard the false ACK and contend again.

[0105] In some examples, when all NQLs hop together, there may be ambiguity in a more distant proxy granting contention to a non-existent local endpoint. As such, the non-targeted proxy receives the Sync at a chip timing based on the relative chip offset between the two proxies. Filtering on a low frequency offset at the proxy mitigates the ambiguity. Specifically, the ambiguity may be mitigated by the endpoint closing the frequency loop based on a timing offset of the target proxy. A maximum Doppler is specified. The wrong proxy may jump to a new frequency after a timeout to start a new round of contention which does not include the particular non-existent local endpoint.Local-to-Proxy (L2P) Communication

[0106] Referring now to Figure 14, a method 1400 of local-to-proxy (L2P) communication is shown, according to an example embodiment. At step 1402, in the U phase, an endpoint may send multiple PDUs. A proxy may be able to perform a process opposite to forward error correction (referred to as “de-FEC”) when a sufficient number of PDUs are received. U may be on a Gold Code specific to a particular proxy but common across all local endpoints for that particular proxy.

[0107] Because there exists an active session with a local endpoint, at step 1404, the transmission (TX) frequency is LOCKED for the duration of the local endpoint communication. Beneficially, locking the TX frequency may allow the strongest local endpoint to maintain its relatively strong channel. At step 1406, the active local endpoint may continue to transmit PDUs in the U slot until there is a Chip Offset 0 sent in the D slot. The294911-8131-1070.1Atty. Dkt. No. 138645-0120Chip Offset 0 sent in the D slot may signal an ACK that indicates that a sufficient number of PDUs have been transmitted for de-FEC to be performed. At step 1408, responsive to the ACK signal, the frequency is UNLOCKED and fast forwards to the current slot number sequence in the pseudo-random order. Another round of contention occurs with the remaining number of local endpoints to be serviced.

[0108] At step 1410, in an embodiment in which a proxy or endpoint is jammed by a frequency hopping collision, proxies and / or endpoints that have no observed signal may time out after a predetermined number of beats (e.g., 3 beats). Local endpoints may then contend again for the channel, and proxies may jump to a current frequency and support the contention protocol. At step 1412, the proxy continues until it does not identify any contention requests for a predetermined period of time. Responsive to the proxy not identifying contention requests within the predetermined period of time, the proxy goes to sleep for a remainder of the 1 minute slot.Proxy-to-Infrastructure (P2I) and Infrastructure-to-Proxy (I2P) Communication

[0109] Referring now to Figure 15, a method 1500 is shown for proxy -to-infrastructure (P2I) and infrastructure-to-proxy (I2P) communication. In P21 and I2P communication, at step 1502, a proxy may serially roll through the identity of all associated local endpoints. The proxy may then perform, at step 1504, UL / DL communication to the infrastructure. In some embodiments, the round trip distance between a proxy and C is added to the FEC codeword. In one example, this may be allowed without additional processes being performed. In another example, the proxy may wrap an additional layer of authentication.Proxy-to-Local (P2L) Communication

[0110] Referring now to Figure 16, a method 1600 illustrating proxy-to-local (P2L) communication is shown, according to an example embodiment. At step 1602, a proxy sends, at the known Sync timing offset, for the first endpoint to be serviced at the known hopped frequency. Since there is an active session with a local endpoint, at step 1604, the TX frequency is LOCKED for the duration of the local endpoint communication. Based on the spreading factor (SF) signaled in the SF-Sync (with SF codes in the chip delay), the proxy304911-8131-1070.1Atty. Dkt. No. 138645-0120 sends, at step 1604, a UL ACK PDU. The proxy may also send DL PDUs (if present) at the unique Gold Code of that endpoint based on its Short ID.

[0111] At step 1606, the active local endpoint then continues to receive PDUs in the D slot. The local endpoint receives the PDUs until the endpoint sends a UL PDU that indicates ACK of both the UL SDU ACK and the DL SDU ACK. After this UL PDU is sent, at step 1608, a next or subsequent endpoint is notified, based on the Sync from its prior L2P, that it is to communicate with the proxy. In an embodiment in which a proxy and / or endpoint is jammed by a frequency hopping collision, at step 1610, proxies and / or endpoints that have no observed signal time out after a predefined number of beats (e.g., 3 beats). Endpoints may reattempt listening beginning at the next minute. However, the endpoints may reattempt listening time at the frequency hopped frequency in the event that the L2P frequency has gone bad. In some embodiments, the process may revert to the contention process of L2P described above. The proxy may continue until, at step 1612, at least one of (a) all ACKs are received or (b) the proxy does not see any contention requests for a predetermined period of time. Upon one or both of (a) or (b) occurring, the proxy may go to sleep.

[0112] The hopping sequence described herein may be a hash of a proxy’s NLQ and SECONDS in the hour (common with L2P), and may also include the MINUTE PHASE. This may cause the P2L to remain on an uncorrelated frequency hop pattern with L2P and other P2L transactions. Thus, only P2Ls of a given NLQ with the same MINUTE PHASE hop together.

[0113] A non-limiting example timeline is described herein. The example timeline begins at 5am. At each hour, including at 5 am, the endpoint demodulates the HOUR Mapping. At around 5:30 am, an event happens that generates an asynchronous SDU. The endpoint may know or determine, from the demodulation at 5 am, that the next opportunity to transmit is at 5:47 am. Further, the endpoint may determine that the next opportunity to transmit is at NLQ 3. This may be a low density scenario. As such, the endpoint may then sleep until MINUTE 47, when the next opportunity to transmit occurs. Table 1 below shows events occurring at various beats in accordance with the example timeline.314911-8131-1070.1Atty. Dkt. No. 138645-0120Table 1

[0114] Continuing the example, timeline, the endpoint demodulates the 10 MINUTE ENCODING SYNC and finds the expected Sync energy corresponding to NLQ 3 at Beat 9410, shown in Table 2 below. Starting at Beat 9411, contention for the proxy begins where the local endpoint, referenced as L2 in Table 2 below, is competing with another local endpoint, referenced as LI in Table 2 below, for the opportunity to send its UL SDU. Since LI has the stronger channel, LI wins the contention against L2 and transmits its UL SDU. L2 continues to look for other options by demodulation of the current frequency. L2 may not see any sync energy until 9414, when the proxy hops onto the current frequency. This time, contention is successful based on L2 seeing its GRANT SYNC at its timing offset. From Beat 9416 through 9418, LI sends its UL SDU until the ACK-PDU is received. At this point, LI goes to sleep.324911-8131-1070.1Atty. Dkt. No. 138645-0120Table 2

[0115] As shown in Table 3, the proxy is successful in transmitting both UL SDUs to the network and receives a DL SDU. The satellite operation may take a duration of time to complete (e.g., a few minutes). As such, the satellite operation may complete around 5:49 am. Thus, at 5:50 (e.g., a MINUTE boundary) the proxy sends a CONTENTION INVITE SYNC on its current hopped frequency. LI and L2, which are monitoring for the CONTENTION334911-8131-1070.1Atty. Dkt. No. 138645-0120INVITE SYNC at the MINUTE rate, may identify or receive the signal. Both LI and L2 may then monitor for a GRANT SYNC at the hopped frequency based on the proxy’s known MAC ID. In this case, LI sees the Grant Sync at Fl, and transmits a Sync that has timing delay corresponding to a SF selection. The proxy then sends the UL SDU ACK PDU at that SF, and LI sends the UL SDU ACK- ACK notifying the proxy that LI is done and going offline. When L2 hears or receives the CONTENTION INVITE SYNC at F2, L2 follows a same or similar procedure as that described with reference to LI. However, in the process performed by L2, the process may include the additional step of receiving its DL SDU. When the UL SDU ACK- ACK and the DL SDU-ACK are received, the proxy determines that the processes to be performed by the proxy are complete and goes to sleep.344911-8131-1070.1Atty. Dkt. No. 138645-0120Table 3

[0116] Table 4 below shows physical building blocks of the system described herein.354911-8131-1070.1Atty. Dkt. No. 138645-0120364911-8131-1070.1Atty. Dkt. No. 138645-01204911-8131-1070.1Atty. Dkt. No. 138645-01204911-8131-1070.1Atty. Dkt. No. 138645-0120394911-8131-1070.1Atty. Dkt. No. 138645-0120Table 4P2L Paging Operation

[0117] Figure 17 provides additional details on the power-efficient paging operation, illustrating the operation between a local function and a proxy function. Specifically, Figure 17 is a protocol diagram showing a method 1700 for power-efficient paging operation between Local and Proxy Functions. The flow of the method 1700 may be such that a proxy serves a local endpoint or communicates with infrastructure with a relayed paging channel for endpoints that do not have the link quality to listen to the paging channel directly. Further, in the Unicast PHY receive, only demodulation of the strongest link is assumed. Therefore, new MAC multiple-access elements may be included. Further, to optimize capacity, the peerpeer communication is frequency hopped.

[0118] The method 1700 begins at step 1702. Specifically, at step 1702, the local function begins in an unassociated state when seeking a proxy based on weak or no satellite connectivity. This may lead to an Hourly Monitor state at step 1704. At step 1704, the endpoint demodulates sync symbols at the beginning of the hour via the frequency offset raster (FOR) to encode the minute. The endpoint specifically demodulates the sync symbol with a sufficient link quality and learns the minute it is available via the FOR.

[0119] The proxy function starts with, at step 1706, a Proxy Battery Budget Available state. Specifically, step 1706 indicates that the proxy has a battery budget available to assist with the local function. The proxy function then transitions to an Hourly Advertisement state at step 1708. The proxy function transitions to step 1708 when battery budget is available to404911-8131-1070.1Atty. Dkt. No. 138645-0120 assist with local function operations. At step 1708, an endpoint selecting, pseudo-randomly, a minute to be available. The minute may occur in the next hour. This may be signaled at the beginning of the hour via the sync sequence, with the FOR to encode the minute. The two functions (e.g., the local function and the proxy function) then converge through a series of synchronized states. For example, at step 1710, the local function participates in the contention request procedure. The local function participates in the request at the encoded minute. At step 1712, the proxy function participates in a contention grant procedure. Specifically, the proxy function participates in the grant at the encoded minute.

[0120] At step 1714, the local function participates in a page registration request. Specifically, the local function learns a MAC ID to predict a frequency hopping pattern. The local function then requests that its paging cycle and / or paging ID be monitored by the proxy function. At step 1716, the proxy function participates in a page registration grant and monitors the local function’s paging cycle and / or paging ID. The network paging cycle and paging ID are conveyed to the proxy to monitor on the local function’s behalf.

[0121] At step 1718, the local function monitors at the same slotted paging cycle that Is set by the network. The local function monitors at a frequency hop frequency based on the proxy function MAC ID. After M number of consecutive paging sync demodulations with no success, the local function returns to an unassociated state (e.g., the method 1700 returns to step 1702). At step 1720, the proxy function participates in a proxy paging cycle with the network. Specifically, the proxy function monitors the local paging slot of the network. If the proxy function matches the paging ID of the local function, the proxy forwards the page to the local function via a sync sequence. After M number of consecutive paging sync demodulations with no null paging response, the method 1700 returns to step 1706.

[0122] At step 1722, the local function enters into at least one of the P2L state, the P2I state, or the L2P state. At step 1724, the proxy function also enters into at least one of the P2L state, the P2I state, or the L2P state. If N number of slotted pages do not result in a page, the proxy may inter a null page as a keep-alive.414911-8131-1070.1Atty. Dkt. No. 138645-0120

[0123] The feedback channel described above (e.g., steps 1710 through 1724), which includes the paging channel, plays an important role in the system. The endpoint monitors the feedback channel at a decimated rate to conserve power, while still receiving necessary updates and commands. This approach allows the system to maintain communication efficiency while significantly reducing the endpoint's power consumption during idle periods.

[0124] In various examples, the method 1700 may include a variable number of syncs based on channel conditions. In addition, the method 1700 may include a network paging reregistration. This may align the proxy and local endpoint slotted paging cycle to power consumption efficiency. The proxy may only need to receive satellite transmission on a single paging cycle to support both the proxy and the local endpoint.

[0125] Table 5 below shows an edge case analysis.424911-8131-1070.1Atty. Dkt. No. 138645-01204911-8131-1070.1Atty. Dkt. No. 138645-0120Table 5

[0126] Figures 18 and 19 illustrate the frame structure used in the small cell ultra-low power side channel protocol.

[0127] Referring specifically to Figure 18, a block diagram 1800 is shown illustrating a fairness implementation. In such an implementation, endpoints may construct a metric, represented as a joule count (J) of a proxy’s role, divided or normalized by its data usage (D) in its local endpoint role. This may result in a parameter referred to as an activity threshold (AT). As shown in Figure 18, a filtered “proxy” joule count 1802 is used, in addition to filtered “local” data usage 1804, to determine an activity factor state variable 1806. The proxy joules may be normalized b local data usage, which may slowly module the activity factor444911-8131-1070.1Atty. Dkt. No. 138645-01201806. A gain 1808 from an infinite impulse response (IIR) filter (e.g., 1-a) may also be used to determine the activity factor 1806.

[0128] A proxy’s unique MAC ID, which may be, for example, 40 bits, and a number of minutes since a last epoch may be inputs into a hash operation 1810. The hash operation 1810 may occur, for example, every minute. The hash operation may be or include a uniform distributed pseudo random number. The output of the hash operation 1810 may be a random challenge (RC). The random challenge may be compared to the output activity factor 1806 at the compare step 1812. When a random challenge is such that the value is less than a current activity threshold (e.g., RC < AF), the random challenge may begin activity as a proxy at the one minute tick. That is, the endpoint may be active as a proxy for the current or next minute. In various examples, the activity threshold may be updated each day based on a filtered J / D parameter value. In one example, an activity factor of 45 may correspond to proxy available once per day averages overtime (e.g., 65536 / 1440 minutes per day = 45.5). In another example, a repeater is a degenerate case of “always available.” Thus, the activity factor may equal 65536.

[0129] In some embodiments, a proxy endpoint may perform a round trip range measurement on sync symbols transmitted by challenged endpoints. A proxy endpoint may then package the measurement value with a MAC ID in the UL SDU to the infrastructure. The value and the MAC ID may be packaged in a dedicated frame format. The proxy endpoint may then FEC the SDU with a field including the packaged measurement value and MAC ID. In various embodiments, a position of a satellite or object may be determined when a plurality of proxy endpoints package information and transmit the SDU. For example, three transactions, each with a different proxy endpoint having a location known to the network, may be performed to determine a position. In various examples, greater or fewer transactions may be needed to determine a position. For example, when a distance between a proxy and a C is less than a predefined distance, a location may be determined in a contention resolution,454911-8131-1070.1Atty. Dkt. No. 138645-0120 as the elements needed to perform location determination may be included in contention resolution for super symbol processing.

[0130] In some embodiments, there may exist difficult-to-reach devices in a network. In such embodiments, a vehicle may be equipped with a proxy endpoint that may drive to a location near a challenged or difficult to reach device. The proxy endpoint may then be able to receive the message from the challenged device. In various embodiments, planning for a tower-infrastructure may use a known location of a proxy and challenged endpoints.Small Cell Ultra-Low Power Side Channel Protocol for Location

[0131] Referring specifically to Figure 19, a diagram of a frame structure 1900 for a communication system is shown, according to an example embodiment.

[0132] Maximum performing location tracking (e.g., location tracking that optimizes battery and latency) may depend upon on laws of physics for each mode of connection. Specifically, for small cell location tracking, a “side-channel” protocol may be used or implemented to minimize endpoint power consumption while maintaining low latency.

[0133] For example, geofences may enter into a tracking mode to track a location. This may quickly deplete a battery. However, the protocol discussed herein with respect to Figure 19 may have a power profile dominated by a 1 sync symbol TX for example, every 10 seconds (for ~10 second response time). This protocol may result in reduced or minimal average power consumption. This average power may be greater than power capabilities of various existing GPS systems.

[0134] Further, the complexity of the protocol described herein may continuously update in the cloud. The protocol allows for accurate round trip ranging (to, for example, 5 meter accuracy) and signal strength. This data can be fused with other sources of data, such as building surveys, for increased indoor positioning accuracy.464911-8131-1070.1Atty. Dkt. No. 138645-0120

[0135] The frame structure 1900 shown in Figure 19 consists of multiple components organized to facilitate efficient data transmission and reception. The frame structure includes a series of page slots 1902, including a page slot zero 0, a page slot one 1, a page slot three 3, and a page slot four 4. These page slots 1902 are arranged vertically within a 10-second time period, allowing for regular paging operations. The frame structure 1900 may be configured such that one slot in a certain number of slots (e.g., one in every five slots) is dedicated to location. The paging subslot may be able to convey small amounts of data (e.g., 5 bits). The frame structure 1900 incorporates 32 sync symbols between the page slots and location groups to maintain synchronization. The location groups 1904 are divided into 16 x 32 symbol subslots at the bottom of the structure 1900, allowing for organized data transmission. In some embodiments, the frame structure 1900 may include one or more additional strips of paging channels.

[0136] Horizontally arranged within the frame structure are location groups 1904, numbered from zero to seven. Specifically, a downlink slot is divided into location groups 1904. This may allow an endpoint to resolve additional small cells in the presence of a strong dominance small cell, which may be required to support multilateration. The location groups 1904 include a location group zero, a location group one, a location group two, a location group three, a location group four, a location group five, a location group six, and a location group seven. Each location group 1904 may have a certain length (e.g., a 50 symbol length). This may allow for a timing error. For example, a length of + / - 9 symbols may correspond to a timing error of + / - 18ms while maintaining a 32 symbol overlap. Each location group 1904 further contains a number of symbol subslots 1906 that facilitate regular communications over an 8-second period. As shown in Figure 19, each location group 1904 includes two symbol subslots. The subslots 1906 on the uplink portion of the location groups 1904 are used to minimize confusion of the identity of the endpoint participating in the round-trip-ranging protocol.474911-8131-1070.1Atty. Dkt. No. 138645-0120

[0137] In addition, the frame structure 1900 may include tunable gaps between symbol subslots for frequency retune. The frequency may be different for each location group 1904 and the uplink subslots may be the same frequency as the corresponding location group frequency for that small cell. In this way, simultaneous uplinks for all small cells can be supported. The extreme version of a very strong near-far may mean that endpoints may not often transmit to a distant small cell when the serving cell has a strength above a certain threshold value (e.g., -40 dBm or stronger). If endpoints did transmit to a distant small cell, a jam may occur across frequencies. Further, transmission to a distant small cell may not be needed because, at that signal strength, the location may be known to be close to the serving small cell.

[0138] The operational flow of the small cell ultra-low power side channel protocol is depicted in Figures 20 and 21. Specifically, Figure 20 shows a method 2000 for the small cell ultra-low power side channel protocol. The method shows a flow between an endpoint 2002, a small cell 2004, and a cloud 2006 (e.g., a network). The method begins at a “join” step 2010 where the small cell 2004 joins the cloud network 2006. The endpoint 2002 may then join the small cell 2004. The cloud 2006 may allocate a geographically specific hopping parameter H to the particular small cell 2004 that joins the cloud network 2006. The hopping parameter H may be a number between 1 and 256.

[0139] The process continues with the endpoint 2002 receiving, at a “join” step 2012, paging information (e.g., a paging index and / or paging slot) and initial location configuration parameters from the cloud system 2006 through the small cell network 2004. This protocol may achieve a lower average power consumption compared to GPS in indoor environments. As such, the method 2000 may achieve an improvement (e.g., a 500x improvement) in power efficiency for location tracking.

[0140] Following the initial configuration, the endpoint 2002 performs a location search at step 2014. During step 2014, the endpoint 2002 demodulates all location group syncs to484911-8131-1070.1Atty. Dkt. No. 138645-0120 multiple cells for position determination. For example, the endpoint 2002 demodulates all location group syncs (e.g., 4 syncs). The demodulation may be performed at a configurable periodic rate to identify a sufficient set of location groups to participate in multiliterate. In various embodiments, this may include a full 32 symbol demodulation as a result of a timing recovery.

[0141] This demodulation at step 2014 may allow the endpoint 2002 to learn timing and a hopping index for multiple small cells 2004 based on a frequency offset rater (FOR). The system utilizes a fine frequency offset raster (FFOR) with a 120 Hz resolution to distinguish between different small cells 2004. This process allows the endpoint to gather information from various small cells in the vicinity while maintaining low power consumption through selective scanning of frequency rasters. The endpoint 2002 may demodulate a full range of frequencies, since the FOR may be signaled in the frequency offset. The FOR may encode a value of, for example, 1-32 to encode the hopping sequence of the small cell 2004. The encoded value may be an input into the formula: V = (F_rast / 5 kHz)+32 x (Location Group number). In various embodiments, the protocol described herein may support resolution of multiple small cells 2004, as they are at different frequencies and are resolved with a maximum detect metric per frequency hypothesis.

[0142] After the location search at step 2012, the protocol initiates a Round Trip Ranging (RTR) procedure at step 2016. The endpoint 2002 transmits or responds to a particular small cell 2004 in the subslot and master raster timing offset based on MAC ID statistics (e.g., a hash). The identity of the particular endpoint 2002 is signaled by the selection of the subslot and the fine frequency offset raster (FFOR) with, for example, 16 kHz / 120 Hz = 130. This yields a space 16 subslots (SS) x 130 FFOR = -2000 TAGs to disambiguate endpoint RTR responses. The selection of the TAG is based on a hash of the endpoints MAC ID. In various embodiments, the polarity of the frequency loop may compensate for timing. Without Doppler, the reception of FFOR may be unambiguous. If Doppler is greater than a predefined494911-8131-1070.1Atty. Dkt. No. 138645-0120 value, such as, for example, 16 mph (60 Hz), the range of valid FFOR may be increased or opened.

[0143] The Round Trip Ranging (RTR) operation at step 2016 includes a low power receive mode where the endpoint 2002 transmits a specific identity signal calculated as V = (F_rast / 5 kHz) + 32 x (Location Group number). This formula enables precise disambiguation between multiple endpoints transmitting simultaneously. The system accommodates, for example, + / -9 symbols or + / - 18 ms timing error allowance during this process.

[0144] In various embodiments, the number of Sync Symbols may be dictated by channel conditions. In some examples, the number of Sync Symbols may be one symbol if within a predetermined distance from edge of cell conditions (e.g., within 13 dB of edge of cell conditions). This may result in reduced power consumption.

[0145] The method 2000 continues with a round trip range and signal strength step 2018. The cloud 2006 may map the round trip range measurements to particular endpoints 2002. If there exist multiple measurements of the same subslot (SS) and timing offset raster (TOR) combination, the cloud 2006 may disambiguate based on position continuity. For example, from the perspective of the endpoint 2002, there may be ambiguity between endpoints. For example, multiple endpoints may have the same Subslot (SS) and FFOR. In such a situation, continuity of location (e.g., chip timing), and continuity of residual frequency offset (e.g., frequency subtracting out 5 kHz raster quantization) can be used to associate a measurement correctly. From the perspective of a small cell 2004, endpoint measurements may be beyond a particular range. For example, endpoint measurements may be outside of the range |d 1 - d2| to dl+d2 of known small cell measurements. Endpoint measurements outside of this range may be disambiguated from being related to a particular endpoint of interest.

[0146] Based on the cloud 2006 receiving multiple measurements with both an accurate time-of-flight and signal strength, location fixes and geofence violations are signaled to the504911-8131-1070.1Atty. Dkt. No. 138645-0120Application. This may also include side information available to the cloud 2006, such as RF surveys of indoor locations. The cloud 2006, based on this information, can indicate to the endpoint 2002 to adjust parameters governing its rate of round trip ranging, location search, and paging cycle. For example, on a geofence trigger, these update rates can be increased.

[0147] The protocol may then incorporate a paging demod step 2020. At step 2020, at the assigned paging slot, the endpoint 2002 demodulates sync if the master raster timing offset matches. If a paging index is match, the endpoint 2002 completes paging slot demodulation and extracts S bits of command. Round trip ranging repeats in multiples of 10 seconds, while location search occurs less frequently than round trip ranging. The paging demod maintains the same rate as round trip ranging. The multiple access RTR protocol enables numerous endpoints to perform ranging operations simultaneously through careful subslot and frequency raster calculations.

[0148] The cloud system 2006 plays an important role in the protocol's operation. Based on the collected data, the cloud 2006 decides to increase or decrease rates of round trip range and location determination based on endpoint movement. This approach optimizes power consumption while maintaining accurate location tracking. The system implements an Over- The-Air (OTA) timing correction mechanism that continuously adjusts endpoint timing to maintain synchronization with the network.

[0149] The interaction between components in the small cell ultra-low power side channel protocol involves a continuous feedback loop. The endpoint 2002 transmits location data and receives configuration updates through the small cell network 2004. The small cells 2004 act as intermediaries, relaying information between the endpoints 2002 and the cloud system 2006. The RTR response signaling includes specific identification parameters that enable the system to track individual endpoints 2002 even in dense deployment scenarios.514911-8131-1070.1Atty. Dkt. No. 138645-0120

[0150] The cloud system 2006 processes the received data, calculates position fixes, and determines if any geofence triggers have occurred. Based on this analysis, at a configuration update step 2022, the cloud system 2006 sends updated configuration parameters back to the endpoints 2002 through the small cell network 2004. These parameters control the frequency of location searches and round trip ranging operations, allowing the system to adapt to changing conditions and movement patterns. The OTA frequency correction mechanism continuously adjusts endpoint frequency references to maintain precise synchronization with the network. At step 2022, the cloud 2006 may determine whether to increase or decrease rates of round trip ranging and / or paging demodulation, and may determine whether to perform a location search in response to movement. The configuration update may be passed to the endpoint 2002 through a paging channel command at step 2024.

[0151] The protocol ensures efficient use of the transmission rate by dynamically adjusting the frequency of location updates based on endpoint movement and signal conditions. This approach minimizes unnecessary transmissions, thereby reducing power consumption and extending battery life. The downlink and uplink slot organization optimizes the balance between communication reliability and power efficiency.

[0152] In areas with strong signal conditions, the protocol leverages the advantaged transmission energy to achieve rapid and accurate location fixes. Conversely, in challenging environments, the system utilizes the proxy transmission energy to maintain connectivity and location tracking capabilities. The dedicated location slots in the frame structure ensure that location tracking operations do not interfere with regular communication functions.

[0153] The small cell ultra-low power side channel protocol provides a robust and efficient method for location determination while minimizing power consumption. By leveraging the specialized frame structure and adaptive operational flow, the protocol enables accurate tracking in various environments while optimizing battery life for endpoint devices.524911-8131-1070.1Atty. Dkt. No. 138645-0120The system achieves significant power savings compared to traditional GPS-based solutions, particularly in indoor environments where GPS signals are weak or unavailable.

[0154] Figure 21 shows a system 2100 for RTR identity signaling. The cloud 2006 may ultimately decide to map a round trip measurement to both the endpoint 2002 and the small cell 2004. The cloud may perform operations such that this mapping is probabilistically correct with minimal signaling overhead. To that end, Figure 21 describes the scheme to perform such a Disambiguation. On the endpoint side, a time advancing scheme 2102 based on a mutually understood MAC ID results in a Subslot Number (SS) and Fine Frequency Offset Raster (FFOR) that pseudo-randomly jumps over time. The FFOR may be, for example, 120 Hz resolution (coarse frequency offset) and spread out over, for example, + / -8 kHz. The FFOR may spread out the signal in the frequency dimension, and a source of endpoint identification (based on continuity of frequency). As shown in scheme 2104, the cloud 2006 attempts to assign a geographically unique Hopping Assignment (H). If the cloud 2006 incorrectly assigns a hopping assignment, aH-collision may occur. From the perspective of the endpoint 2002, a stronger small cell 2004 may always hide a weaker small cell 2004.

[0155] In some embodiments, NTP may calibrate frequency (aging) over long periods of time (e.g., days, weeks, etc.) for non-satellite connected small cells. In other embodiments, endpoints send SDUs to the cloud, noting large frequency offsets relative to endpoint timing. The cloud may then determine whether a small cell or an endpoint (e.g., via SDU) should correct the frequency offsets.Round Trip Ranging (RTR) Multiple Access

[0156] Referring now to Figure 22, a method 2200 for RTR multiple access is shown, according to an example embodiment. In various embodiments, the small cell may resolve multiple RTR responses in a given subslot. Described herein are processes for a round trip ranging multiple access protocol.534911-8131-1070.1Atty. Dkt. No. 138645-0120

[0157] At step 2202, endpoints may transmit at maximum power to allow a small cell to identify or determine a received signal strength indicator (RSSI) for input into the location algorithm. At step 2204, endpoints may minimize the number of Sync transmissions as well as randomize the symbol timing. For example, if one sync symbol is to be transmitted, then a random draw may be performed for between 1 and 32. If two sync symbols are needed or transmitted, a random draw may be performed for between 1 and 16, etc. This processes may be similar to sub-slot selection on satellite uplinks described herein.

[0158] At step 2206, the endpoint may then implement a dither (e.g., a ~8 kHz dither) in frequency to maximize a probability of simultaneous demodulation while not breaking CPU processing timing for back to back subslot processing. Subsequently, at step 2208, a small cell may analyze the various number of symbol combinations (e.g., aligned powers of 2).Paging Subslot Processing

[0159] Referring now to Figure 23, a method 2300 of paging subslot processing is discussed, according to an example embodiment. To begin the method 2300, every slot may begin with a paging opportunity for each small cell. At step 2302, endpoints are assigned a paging slot (PS) that the network expects it will monitor. The timing offset shift is the address for a particular page which maps uniquely to an endpoint. Further, there may be, for example, 5 bits encoded in each page. In various embodiments, chip spacing at or below a predefined number of chips (e.g., five chips or less) may be resolvable. This may create a space to uniquely address, for example, 1024 / 5 = -200 endpoints per paging slot. At step 2304, endpoint processing occurs. Endpoint processing may include learning, by the endpoint, that the endpoint is not being “paged” with just 2 symbols of demodulation, with the flexibility to stay active longer under near edge-of-cell link conditions. A byproduct of this power consumption optimization may be that there needs to be a “null page” (i.e., a timing offset of 0) when there is no active page so endpoints can abort processing.544911-8131-1070.1Atty. Dkt. No. 138645-0120Round Trip Ranging (RTR) Response Signaling / Identification

[0160] Referring now to Figure 24, a method 2400 for round trip range response signaling and identification is shown, according to an example embodiment. At step 2402, bits can be signaled in the RTR. This may be, nominally, five bits, for example. At step 2404, a subslot and modulation symbols are determined based on the bits to be signaled. As one example, for a single sync symbol, the selection of which of the 32 symbol locations for TX indicates 5 bits. As another example, for a 32 sync symbols, a Hadamard modulation encodes 5 bits. This may be applicable for any power of 2 in between. For example, an 8 sync symbol RTR response signals 2 bits via selection of 4 possible symbol starting locations, and 3 bits via Hademaard modulation. At step 2406, a deterministic scramble is performed on these bits to allow for uniform distribution.

[0161] At step 2408, one or more bits may be spent. At step 2408, various possible spends of these bits may exist. For example, there may be an acknowledgement of configuration updates in the Paging channel. There may also be signal strength signaling. The endpoint can power control for better multiple access of RTR responses and not lose the signal strength information that may be useful for location determination. As another example, more bits may be used to disambiguate which endpoint is the source of the RTR response.Over the Air Protocol

[0162] Over the air (OTA) protocols are discussed herein, particularly with respect to Figures 25-27. Specifically, a firmware over the air (FOTA) distribution system implements three primary modes of operation within an outer loop control framework: satellite broadcast, synchronized peer-to-peer, and incidental peer-to-peer. Figures 25 and 26 illustrate the network architecture for satellite communication and firmware distribution, while Figure 27 depicts the three-phase frame sequence used in the synchronized peer-to-peer mode.554911-8131-1070.1Atty. Dkt. No. 138645-0120

[0163] An OTA timing correction mechanism is discussed herein. Because network timing can be significantly off (e.g., by + / - 100 ms), signaling may be used to pull in timing. Endpoints may be used to help perform such a timing correction. An endpoint that has a link to multiple small cells (or satellites, towers, repeaters, etc.), including a small cell with significant error, may be required to perform this timing correction. If no such endpoint exists, then the small cell being significantly off in timing may have minimal impact and may be disregarded.

[0164] Based on the magnitude of a large timing error, there exists ambiguity of which Location Group is being targeted, since Location Groups may be a small distance (e.g., 100 ms) apart. Further, a certain number (e.g., half) of the symbols may be in the 32 symbol window of a time-locked small cell with, for example, a 3 dB degradation in sensitivity. Thus, it may be beneficial to lock the timing of this errant small cell as soon as possible. If a signal is found in a Location Group, nominally that small cell FOR and Location Group in the future may be predictable. However, given that there is a 3-fold ambiguity (e.g., on-time, early, and late) in the Location Group Slot, the endpoint may look in 3 places for the small cell in the next location operation to understand the errant small cell timing error Location Group alias. With this information, the endpoint can send an SDU to the cloud with the full error that would likely be in the -50 to +50 symbol range. If the cloud believes that the reference for that measurement is valid (i.e., a timing locked small cell), the cloud may command a timing offset update to the errant small cell.

[0165] An OTA frequency correction mechanism is discussed herein. In some examples, an aged small cell or endpoint may come on-line with, for example, >2.5 kHz (1 ppm) error. To correct the frequency, the endpoint may perform a hypothesis test. Nominally with a Frequency Offset Raster (FOR) and Location Group, the future of that small cell may be predictable. If that small cell is not found in the expected location, then an adjacent Frequency Raster Offset’s (FOR) predictive ability may be performed. If a multiple of 5 kHz offset is564911-8131-1070.1Atty. Dkt. No. 138645-0120 determined, the determination is signaled to the cloud, and the cloud may determine which end of the link (e.g., small cell or endpoint) should correct timing. In another example, during steady-state aging, signaling may occur prior to achieving a threshold value (e.g., a +Z-2.5 kHz threshold) that causes FOR ambiguity. In some examples, there may be a joint significant Timing and Frequency error. In such an example, the endpoint may explore a 2-d ambiguity space of Location Group error and FOR space. One manner of decoupling may be for an endpoint to note the frequency offset of the Paging Symbols (which may be sent with no intentional carrier offset) and signal when the frequency offset is out of bounds. In many cases, the cloud may be able to resolve whether the error results from an errant endpoint or an errant small cell

[0166] The firmware over the air (FOTA) distribution system implements three primary modes of operation within an outer loop control framework: satellite broadcast, synchronized peer-to-peer, and incidental peer-to-peer. Figures 25 and 26 illustrate the network architecture for satellite communication and firmware distribution, while Figure 27 depicts the three- phase frame sequence used in the synchronized peer-to-peer mode. Specifically, Figure 25 shows a firmware over the air (FOTA) system 2500. As shown, the FOTA system 2500 includes a satellite 2502 and a plurality of endpoints 1-8. The satellite 2502 may communicate with endpoint 1 and endpoint 2 through satellite P2P communication. Satellite P2P communication may be global in nature and may occur during favorable link conditions (e.g., the endpoints are not obstructed and able to be reached by the satellite 2502). Satellite P2P communications may be capable of a low spreading factor (e.g., 64) firmware download from the satellite 2502.

[0167] In the satellite broadcast mode, a satellite 2502 transmits consecutive data PDUs to multiple endpoints (e.g., endpoints 1 and 2 in Figure 25). The transmission rate may be, for example, % FDU with PDUs delivered in sequence. The satellite 2502 utilizes the transmission rate to efficiently distribute firmware updates across a wide coverage area.574911-8131-1070.1Atty. Dkt. No. 138645-0120Endpoints with direct satellite visibility receive the firmware updates at the advantaged transmission energy. That is, many, but not all, endpoints may be able to receive Comm / AP FW through the satellite broadcast mode. This mode employs Reed-Solomon algorithms to reconstruct missing systematic data when additional parity bytes are received, enabling efficient recovery of complete firmware images without requiring retransmission of specific missing segments.

[0168] As shown, there may be challenged link conditions between the satellite 2502 and each of endpoint 3, endpoint 4, and endpoint 6. In such challenged link conditions, the satellite 2502 may be incapable of providing a low spreading factor firmware download. Accordingly, synchronized P2P communication may occur. A synchronized P2P communication mode may be coordinated by the network to aggressively target those that were missed during the satellite broadcast mode (e.g., satellite P2P communication). The synchronized P2P mode allows for finding the nearest FT (FOTA transmitting) globally by concentrating the action to a particular time, which optimizes battery draw. The synchronized P2P mode also allows for FT to service multiple FR (FOTA receivers) simultaneously, which also optimizes battery draw.

[0169] Synchronized P2P communication may have a transmission rate of, for example, / i FEC, with PDUs sent in a randomized order. This may enable or support continual handovers for a FR receiving data from multiple FT s that are not data synchronized. In various examples, only endpoints that are FT for a particular image make their presence known via Sync Symbols at a predictable hopping pattern. Further, synchronized P2P communication may be governed by rules that minimize required power. For example, FRs may request only the strongest available FT to minimize SF. Further, FRs may only request when the channel is relatively strong to take advantage of constructive fades to minimize SF. Synchronized P2P communication may occur at a set time of day (i.e., midnight GPS time for Comm FW) and other times for Application FW.584911-8131-1070.1Atty. Dkt. No. 138645-0120

[0170] Specifically, as shown in Figure 25, synchronized P2P communication may occur between endpoint 1 and endpoints 3 and 4. Synchronized P2P communication may also occur between endpoint 2 and endpoint 6. Further, as shown, synchronized P2P communication may occur from endpoint 4 to endpoint 5, and from endpoint 5 to endpoint 6. Synchronized P2P communication may provide or allow for a peer-to-peer firmware download.

[0171] In synchronized P2P communication, peers may multicast to multiple endpoints simultaneously (e.g., as shown by endpoint 1 simultaneously casting to endpoint 3 and endpoint 4). Further, in various embodiments, peers that receive a firmware download from a peer may then transmit the image to additional peers. For example, endpoint 4 may provide a firmware download to endpoint 5. Endpoint 5, upon receiving the download, may provide the firmware to endpoint 6. As shown, endpoint 6 may receive firmware from endpoint 2 and / or endpoint 6. An endpoint may transition its source of firmware image to a different peer at any time based on each peer sending the data in a randomized order.

[0172] The synchronized peer-to-peer mode enables coordinated firmware distribution between endpoints at a predetermined time (e.g., midnight GPS time for communication firmware). As shown in Figures 25 and 26, “relevant endpoints” that have been commanded via unicast to participate in the firmware download process monitor and transmit syncs at particular hopping frequencies. Endpoints 1 and 2 with complete firmware images transmit on a specific frequency hopping pattern that changes, for example, every 3 seconds, while endpoints without complete images (such as firmware receiver FR 1, firmware receiver FR 2, and firmware receiver FR 3 shown in Figure 27) search for sync energy from these transmitters. This mode leverages the proxy transmission energy to facilitate firmware distribution in areas with limited direct satellite coverage.

[0173] The system 2500 further includes incidental P2P communication, as shown between endpoint 7 and endpoint 8. Incidental P2P communication may service endpoints that are off network for a predetermined period of time. For endpoints that have been offline594911-8131-1070.1Atty. Dkt. No. 138645-0120 and have missed the focused action during synchronized P2P, FW versions may be compared during normal P2P interactions, and later versions are sent to endpoints that are out of date under certain circumstances. For example, endpoint 8 may be off-network for a predetermined period of time. Endpoint 8 may then determine, through a normal P2P communication, that endpoint 7 has a more recent firmware image than endpoint 8. Endpoint 8 may then receive this updated or more recent firmware from endpoint 7.

[0174] The incidental peer-to-peer mode allows for opportunistic firmware updates during normal peer-to-peer interactions with limited power expenditure from the peer. In this mode, endpoints that have been offline for a period can discover through routine communication that a peer has a more recent firmware image. Peers with more recent version numbers unicast FOTA PDUs in randomized, non-repeating order during the P2L phase. This mode is particularly useful for maintaining up-to-date firmware in areas with intermittent connectivity or for devices that frequently move between coverage areas. The incidental P2P may be appliable for comm FW, but may take time for application FW, since proximity to a device with the same application is required. Further, the FR endpoint may be required to incidentally choose the correct FT proxy since incidental P2P communication may lack the targeting ability available in the synchronized P2P mode.

[0175] Figure 26 shows a frame structure 2600 for synchronized P2P communication. In the synchronized P2P mode, all “relevant endpoints” monitor and / or transmit syncs at a particular hopping frequency at a particular time (e.g., midnight GPS time). “Relevant endpoints” as referred to herein may be endpoints that have been commanded (e.g., via unicast) to participate in a comm FW download. Endpoints that initially do not have the complete FOTA image may search, on a specific frequency hopping pattern, for Sync energy from those that do have the complete FOTA image. Figure 27 illustrates a 3 phase frame sequence 2700 with a predictable frequency hopping occurring every 3 seconds. Unlike the Satellite Broadcast, the FOTA PDUs that are transmitted are transmitted in random ordering.604911-8131-1070.1Atty. Dkt. No. 138645-0120This allows for seamless handover from one peer to another with the relatively small cost of some probability of FOTA PDU duplication.

[0176] As an example, the frame format for FOTA transmissions includes a 2-byte sequence number and an 18-byte payload, allowing for a total firmware image size of 590 kilobytes after rate 1 / 2 coding. Each code block of 128 x 18 bytes includes a 16-byte AES 128 signature and additional headers for length, block sequence number, and firmware image identification. If an AES signature is unable to be used for a block, a P2P request for a particular block may be made. In other embodiments, transmissions may continue to be received for FEC error correction in combination with overwrites of spurious data that reduces the impact of the spurs over time. Further, in some embodiments, a block size may be increased by encoding a value 1 to 3, in addition to 3 bits signaling on the FOR, to allow all images to fit within the block. This may be applied to all PDUs.

[0177] Figure 27 illustrates the three-phase frame sequence 2700 implemented across three subframes 1, 2, and 3, each lasting one second. In Subframe 1, a Firmware Transmitter (FT) 2702 sends initial PDUs with a session request bit. Firmware receivers 2704 (shown as FR 1, FR 2, and FR 3) receive these PDUs based on their respective link conditions, resulting in received PDUs. The FT 2702 transmits FOTA ADV sync sequences to enable channel sounding at a specific frequency (e.g., F0). In various examples, the FR 2704 may scramble the FI PDU ordering and may not repeat earlier sent PDUs until cycling through all PDUs.

[0178] During Subframe 2, the firmware receivers 2704 send a FOTA RESP sync sequence transmission in the portion corresponding to its required spreading factor to close the link. The FRs use uniform frequency spacing to enable resolving multiple receivers simultaneously. The FT determines the spreading factor based on the set of received FOR- signaled SFs during Subframe 1, optimizing transmission parameters for each receiver. For example, FRs 2704 place a randomly selected offset (e.g., 1kHz offset) between, for example,614911-8131-1070.1Atty. Dkt. No. 138645-0120+ / - 30 kHz to provide a number of values (e.g., 60 when a 1kHz is provided between + / - 30 kHz)for resolving multiple FRs 2704.

[0179] In Subframe 3, the FT 2702 may only attempt transmission if the number of received or projected to be received PDUs exceeds a time-varying threshold T(n). The FT 2702 determines a spreading factor based on a set of received FOR-signaled spreading factors during subframe 2. The FRs 2704 synchronize with a spacing factor in a sync sequence signaled at FOR, with N number of frequency Fl PDUs. N may range from 1 to 256. FRs 2704 may only attempt demodulation and accept firmware upgrades if the upgraded spreading factor (e.g., the FOR-signaled spreading factor) is supported based on measured a measured RSSI, ensuring reliable transmission while optimizing power consumption.

[0180] As shown in Figure 27, one or more FRs 2704 send a FOTA Response (FOTA RESP) sync to the FT 2702 with the strongest link in each frame (e.g., using power control). This may cause bouncing between multiple FTs 2702 with the requested SF codes as a modulated frequency offset raster (FOR). The FR 2704 selects a FOR at, for example, 1 kHz spacing so multiple FOTA RESP may be found with the maximum metric per frequency offset described herein. The number of sync signals may be mapped to be sized to the spreading factor to allow the FRs 2704 to transmit a single sync symbol for all but the lowest of the supported SFs. For example, 1 sync symbol for a spreading factor of 4 to 32, 2 sync symbols for a SF of 64, 4 sync symbols for a SF of 128, 8 sync symbols for a SF of 256, 16 sync symbols for a SF of 512, and 32 sync symbols for a SF of 1024, for a total of 58 symbols over all SFs.

[0181] The result of the reception of the set of FOTA RESP sync symbols is a histogram of the number of endpoints requesting a given SF. From this data, the FT 2702 computes the number of predicted received FOTA DATA PDUs as a function of all possible SFs. This may be expressed as: R_P (SF) = (1024 / SF) x sum(FOTA_RESP in SF bins >= SF).624911-8131-1070.1Atty. Dkt. No. 138645-0120

[0182] The SF optimal is then selected to maximize R_P(SF), whose maximum value corresponds to R P optimal. If R P optimal >= T(n), where T is the threshold for frame n, then the FT will transmit at SF optimal. If SF optimal < T(n), the FT will not transmit that frame. The T(n) function may start at a high value and slowly reduce over time. For example: T(n) = 512 x exp(-cn), where the function decays exponentially at a rate of ~1% per minute (over 20 cycles) which yields c = -0.0005.

[0183] The FT 2702 may signal the selected SF as a frequency offset in the sync symbol in the 3rd phase (e.g., subframe 3) so that endpoints can focus on the particular SF. Specifically, no attempt may be made if the SF has a value less than what may be required to reduce battery power.

[0184] If no transmission has occurred after N frames based on the optimal spreading factor being below the threshold T(n), that FT 2702 attempts a TX Subframe 1 at a once per minute rate. Thus, FRs 2704 may be occasionally jammed by stronger links that are not ready to transmit due to being below threshold value T(n). In various embodiments, FTs 2702 may not exceed an absolute Spread Factor maximum Y, which may be signaled in the outer loop.

[0185] Additionally, when endpoints receive a full image, the endpoints may begin sending the sync advertisement. The endpoint will spend an allocated amount of power (P), signaled in the outer loop, in the service of other endpoints. The amount of power may be, for example 300 J. When an FT 2702 has a consecutive numbers of frames in silence, the FT 2702 goes to sleep. The FT 2702 may also go to sleep if a configurable amount of power P was spent in the service of others. The FR 2704 may go down after Z number of frames of silence, even if the full image has not been received.

[0186] The rationale for the above-mentioned processes is described herein. For example, a decimation rate of ’A may allow for normal operations coexistence. Further, a slowly reduced threshold allows more efficient FTs to make progress even if the FTs are not634911-8131-1070.1Atty. Dkt. No. 138645-0120 the strongest link to a given FR. This may allow for broadcast efficiency. Additionally, for a small number of FRs per FT, the FT may transmit only during channel peaks (via frequency hopping diversity) as T is slowly reduced.

[0187] During incidental P2P communication, there may be a limited amount of power spent on the incidental P2P operation from the peer. If the comparison of version numbers yields a peer that is more recent, that peer may unicast the FOTA PDUs in a randomized, nonrepeating order during the P2L phase.

[0188] The outer loop control framework (e.g., the communication FOTA outer loop framework) begins with the network staging the operation 24 hours in advance. This advance notification allows for operational control over the process, including adjustments to power allocation, performance margins, and participant selection. For example, various parameters can be adjusted by the communication FOTA outer loop. These parameters may be communicated the day before in the unicast messaging exchange. Parameters that can be adjusted include, for example, parameters such as power allocated, that are communicated as an outer loop, and a margin X additional for optimal performance. For dense regions, the comm GW may select only a subset at a time to be chosen to participate in the protocol to reduce odds of flooding. In some embodiments, another outer loop could be initially isolating to just repeaters as FT. Auto flipping to FR may be beneficial in some cases, but if too dense, could cause issues. The outer loop may make a decisions of whether auto flipping to FR should occur.

[0189] In addition, endpoints are notified via unicast messaging, including the day start and stop of the transaction. The endpoints also receive a CMAC for the entire firmware image to authenticate the update. For example, if the entire firmware image is matched with a subsequent image, a switchover may be authenticated. The outer loop continuously learns the percentage completion of endpoints and adjusts parameters accordingly.644911-8131-1070.1Atty. Dkt. No. 138645-0120

[0190] During the FOTA process, an endpoint session ID hash is used to manage session IDs and control the distribution of firmware updates. The header field 301, session hash 302, hash field 303, hash sequence 304, and session wrap field 305 shown in Figure 3 are utilized in the feedback channel to efficiently manage the FOTA distribution process.

[0191] The system implements adaptive spreading factor control using the spreading factor increase command received at step 212 and spreading factor decrease command received at step 214 shown in Figure 2. These commands allow for dynamic adjustment of transmission parameters based on link conditions, optimizing the FOTA distribution process for various network environments.

[0192] In cases where rapid termination of transmission is required, the immediate stop hash 218 of Figure 2 is employed. This feature allows for efficient use of network resources by halting unnecessary transmissions when sufficient data has been received.

[0193] The FOTA distribution system adapts to various network conditions, utilizing the indoor transmission rate and indoor transmission energy for challenging environments, such as indoor locations with limited satellite visibility. For application FOTA, the outer loop operates similarly to the communication FOTA, but with the timing offset from GPS midnight to prevent confusion with communication firmware updates or different application firmware images. Further, the communication gateway (e.g., network) can enlist non-recipient endpoints to accept and transmit the image, further enhancing distribution efficiency.

[0194] A number of implementations have been described. Nevertheless, it will be understood that various modifications may be made without departing from the spirit and scope of the disclosure. Accordingly, other implementations are within the scope of the following claims.654911-8131-1070.1

Claims

1. Atty. Dkt. No. 138645-0120WHAT IS CLAIMED IS:

1. A method performed by an endpoint for satellite communication, the method comprising: transmitting a set of protocol data units (PDUs) at different spreading factors with a session request bit; monitoring a feedback channel for a hash matching an endpoint ID and a minimum spreading factor of transmitted PDUs, wherein the hash is included with other data in the feedback channel; and if a matching hash is found, continuing transmission at a signaled spreading factor and monitoring a command packed in the feedback channel.

2. The method of claim 1, further comprising: if no matching hash is found, attempting transmission at different spreading factors.

3. The method of claim 1, further comprising: reacting to a received command by: stopping transmission if the command indicates stop; increasing spreading factor if the command indicates increase; decreasing spreading factor if the command indicates decrease; or continuing transmission if the command indicates no change.

4. The method of claim 3, wherein stopping transmission occurs when determining that a sufficient number of PDUs have been received to reconstruct a service data unit (SDU).

5. The method of claim 1, wherein the hash is based on a minimum spreading factor PDU that was transmitted by the endpoint.664911-8131-1070.1Atty. Dkt. No. 138645-01206. The method of claim 1, wherein the feedback channel is monitored by multiple active endpoints.

7. The method of claim 1, wherein a bit position where the matching hash is found corresponds to a session ID assignment.

8. A satellite communication system, comprising: an endpoint configured to transmit a set of protocol data units (PDUs) at different spreading factors with a session request bit; and a satellite configured to: receive a subset of the PDUs based on link conditions, construct a hash based on an endpoint ID and a minimum spreading factor of received PDUs, and pack the hash with other data in a feedback channel; wherein the endpoint is further configured to monitor the feedback channel for a matching hash and, if a matching hash is found, continue transmission at the signaled spreading factor and monitor a command packed in the feedback channel.

9. The satellite communication system of claim 8, wherein the endpoint is further configured to attempt transmission at different spreading factors if no matching hash is found.

10. The satellite communication system of claim 8, wherein the endpoint is further configured to react to a received command by: stopping transmission if the command indicates stop; increasing spreading factor if the command indicates increase; decreasing spreading factor if the command indicates decrease; or continuing transmission if the command indicates no change.674911-8131-1070.1Atty. Dkt. No. 138645-012011. The satellite communication system of claim 10, wherein stopping transmission occurs when a sufficient number of PDUs have been received to reconstruct a service data unit (SDU).

12. The satellite communication system of claim 8, wherein the hash is constructed based on a minimum spreading factor PDU that was received by the satellite.

13. The satellite communication system of claim 8, wherein the feedback channel is monitored by multiple active endpoints.

14. The satellite communication system of claim 8, wherein a bit position where the matching hash is found corresponds to a session ID assignment.

15. A non-transitory computer-readable medium storing instructions that, when executed by a processor, cause the processor to perform a method for satellite communication, the method comprising: transmitting a set of protocol data units (PDUs) at different spreading factors with a session request bit; monitoring a feedback channel for a hash matching an endpoint ID and a minimum spreading factor; and if a matching hash is found, continuing transmission at a signaled spreading factor and monitoring a command packed in the feedback channel.

16. The non-transitory computer-readable medium of claim 15, wherein the method further comprises: if no matching hash is found, attempting transmission at different spreading factors.

17. The non-transitory computer-readable medium of claim 15, wherein the method further comprises: reacting to a received command by:684911-8131-1070.1Atty. Dkt. No. 138645-0120 stopping transmission if the command indicates stop; increasing spreading factor if the command indicates increase; decreasing spreading factor if the command indicates decrease; or continuing transmission if the command indicates no change.

18. The non-transitory computer-readable medium of claim 17, wherein stopping transmission occurs when a sufficient number of PDUs have been received to reconstruct a service data unit (SDU).

19. The non-transitory computer-readable medium of claim 15, wherein the hash is constructed based on a minimum spreading factor PDU that was received by a satellite.

20. The non-transitory computer-readable medium of claim 19, wherein a bit position where the matching hash is found corresponds to a session ID assignment.694911-8131-1070.1

Images