LoRa uplink channel capacity expansion method and system based on controllable frequency offset

Abstract

This invention belongs to the field of IoT communication technology and provides a method and system for expanding LoRa uplink channel capacity based on controllable frequency offset. The method includes: a gateway allocating a controllable frequency offset Δf to terminals requiring uplink transmission and issuing configurations; the terminal applying Δf to standard LoRa symbols in a cyclic shift manner to generate characteristic signals; the gateway receiving conflicting mixed signals, identifying Δf through delinear frequency modulation and fast Fourier transform, and separating the peak signals into corresponding sub-channel streams based on the Δf value; and performing parallel decoding on each sub-channel stream to obtain the original uplink data from different terminals. This invention virtually expands a single physical channel into multiple logical sub-channels, actively and controllably separating conflicting signals, is compatible with heterogeneous terminals using active RF and backscatter, significantly improves LoRa uplink channel capacity, reduces conflicts and retransmissions, and extends terminal lifespan, making it suitable for high-density IoT terminal deployment scenarios.

Description

Technical Field

[0001] This invention belongs to the field of Internet of Things (IoT) communication technology, specifically relating to a LoRa uplink channel capacity expansion method and system based on controllable frequency offset. Background Technology

[0002] LoRa (Long Range), based on linear frequency modulation (CSS), has become a key technology in the Internet of Things (IoT) field due to its low power consumption and long-range communication capabilities. However, the standard LoRa network architecture has the following inherent defects, making it difficult to support large-scale terminal deployment: Uplink channel physical capacity bottleneck: Standard LoRaWAN gateways support a limited number of uplink channels (e.g., 8), which does not meet the needs of large-scale, dense deployment of IoT terminals. A single physical channel becomes a bottleneck for data aggregation.

[0003] Simple random access mechanism: To reduce power consumption and cost, LoRa terminals typically lack carrier sense capability, employing a simple ALOHA random access mechanism for uplink transmission. The terminal randomly selects a channel to send data, making it highly susceptible to collisions at the gateway.

[0004] Data convergence conflict in star topology: In a typical LoRa star network topology, all data from terminals within the coverage area needs to be sent to the same gateway. When multiple terminals send data simultaneously or nearly simultaneously, their signals overlap in the time and frequency domains, causing a conflict. Standard LoRa gateways cannot separate and decode multiple data packets from such conflicting signals, leading to transmission failure.

[0005] The chain reaction of negative effects triggered by the conflict: The aforementioned conflict directly leads to packet loss, a sharp drop in network throughput, and an increase in end-to-end transmission latency. To ensure reliability, terminals need to retransmit frequently, which significantly increases the energy consumption of terminals (especially battery-powered devices), shortens their service life, and raises network maintenance costs.

[0006] Compatibility challenges with emerging technologies: Emerging ultra-low-power LoRa backscatter terminals often use channels improperly and lack effective capacity expansion mechanisms. Their access further exacerbates uplink conflicts with traditional LoRa networks, and a unified solution is lacking.

[0007] To address the above problems, existing technologies mainly seek solutions from two directions: transmission mechanism optimization and post-collision signal decoding. 1. Collision decoding using the inherent frequency offset (Δf) of the hardware: This approach reveals that due to minute differences in hardware components such as crystal oscillators, the signals transmitted by different LoRa terminals exhibit an inherent, uncontrollable frequency offset (Δf) (CFO). In the event of a collision, the gateway can utilize this hardware "fingerprint" to distinguish signal components originating from different source devices.

[0008] Implementation method: By demodulating and analyzing the received conflicting mixed signals, the unique hardware frequency offset (Δf) carried by different signal components is identified, and then multiple data packets are separated and decoded.

[0009] The limitations of this approach are: The dependence on an uncontrollable dimension: the frequency offset (Δf) utilized is caused by hardware manufacturing defects, and is not actively introduced, unconfigurable, or controllable. Its offset value is random and finite, limiting the number of distinguishable devices.

[0010] Poor flexibility and scalability: It is impossible to dynamically adjust or allocate frequency offset (Δf) resources according to network load, making it difficult to achieve efficient network scheduling and management.

[0011] Limited performance improvement: This method "distinguishes different symbols by using the decimal point of the frequency offset (Δf), but the limited variety of available random frequency offset (Δf) values ​​may limit the performance improvement."

[0012] 2. Other related conflict decoding schemes mLoRa attempts to find partially overlapping segments in conflicting signals and decodes them by iteratively constructing and eliminating symbols. However, it relies on temporal misalignment between conflicting signals and performs poorly in conflicts with highly aligned symbols.

[0013] CoLoRa: Converts the time offset between data packets into measurable frequency characteristics, thereby separating colliding packets. It is also limited by the specific time-domain characteristics of the collision signal.

[0014] FTrack: Separates collisions by exploiting temporal misalignment at the edges of the collision symbols. However, this method is less effective when the collision symbols are highly aligned temporally.

[0015] 3. Transmission Mechanism Optimization Scheme CAD-based CSMA: This approach allows terminals to perform Channel Activity Detection (CAD) before transmitting to assess channel activity and avoid collisions. However, in high-density scenarios, CAD operations themselves incur significant additional energy costs and channel access latency, potentially reducing overall network efficiency.

[0016] Adaptive spreading factor (SF) allocation: SF is dynamically allocated based on factors such as the signal strength (RSSI) of different terminals, utilizing the orthogonality between different SFs to reduce interference. However, this does not increase the channel's logical capacity, and SF resources themselves are limited. Summary of the Invention

[0017] The purpose of this invention is to overcome the existing defects and provide a method and system for extending LoRa uplink channel capacity based on controllable frequency offset.

[0018] To solve the above-mentioned technical problems, the present invention provides the following technical solution: A method for extending the uplink channel capacity of LoRa based on controllable frequency offset includes the following steps: The gateway assigns a unique sub-channel identifier to each terminal that needs uplink transmission. The sub-channel identifier is associated with a predefined controllable frequency offset value. ; The terminal generates standard LoRa symbols based on the data to be transmitted, and applies the terminal subchannel identifier corresponding to each Chirp symbol in a frequency-wrap cyclic shift manner. Generate and send a feature signal carrying a frequency offset fingerprint; The gateway continuously monitors the uplink signals of all terminals and analyzes the mixed conflict signals by delinear frequency modulation and fast Fourier transform. The gateway analyzes each peak in the Fast Fourier Transform spectrum and calculates the signal of each peak. According to different The peak signal is separated into the corresponding sub-channel stream, and each sub-channel stream is decoded in parallel to obtain the raw uplink data from different terminals.

[0019] Furthermore, the frequency offset value Δf satisfies the constraint condition. ,in For channel bandwidth, is the spreading factor.

[0020] Furthermore, the cyclic shift method specifically involves: forcing the instantaneous frequency of the Chirp symbol to jump and wrap back to the opposite boundary when it reaches the channel bandwidth boundary during linear modulation, ensuring that the symbol energy is always constrained within the standard channel bandwidth.

[0021] Furthermore, the terminal includes an active radio frequency terminal and a backscatter terminal; For active radio frequency terminals, software-defined radio is used to directly generate cyclically shifted Chirp symbols during the baseband signal processing stage; For backscatter terminals, Δf is obtained by demodulating the downlink excitation signal command of the gateway, and a local voltage-controlled oscillator is configured to generate a subcarrier with a frequency corresponding to the subchannel. After mixing and reflection, a reflected signal is generated that is located in the standard uplink channel and carries the characteristics of the specific subchannel.

[0022] Furthermore, the gateway dynamically adjusts the Δf value allocated to the terminal via the downlink based on network load and conflict history, achieving adaptive allocation and scheduling of sub-channel resources, specifically including: Monitoring steps: The gateway continuously receives uplink signals, demodulates and identifies each element in the Fast Fourier Transform spectrum. Record the frequency of decoding failure or retransmission corresponding to the value, and record the collision rate of each sub-channel. Decision-making steps: The scheduling management unit analyzes conflict data, identifies load hotspots, and determines the terminals and target Δf values ​​that need to be adjusted; Execution steps: Send a reconfiguration command to the target terminal via the downlink; Feedback steps: The target terminal updates its local Δf value, and subsequent uplink data is sent using the new Δf value.

[0023] Furthermore, the gateway adopts different Δf value adjustment strategies based on different load states, including: Initial allocation state: The gateway adopts a round-robin allocation strategy to allocate a unique, unused Δf to the terminal, eliminating random access conflicts; Conflict occurrence status: The gateway switches the terminals on the high-conflict sub-channel to the currently idle or less conflicted sub-channel to achieve load balancing; In resource-scarce situations: The gateway combines frequency hopping mechanisms and transmission timing control to prioritize the transmission of important data and maximize the decoding success rate.

[0024] Another objective of this invention is to provide a LoRa uplink channel capacity extension system based on controllable frequency offset, comprising: The configuration module, located on the gateway, is used to assign a unique sub-channel identifier to each terminal requiring uplink transmission. Each sub-channel identifier is associated with a predefined, controllable frequency offset value. ; The signal generation module, located in the terminal, is used to generate standard LoRa symbols based on the data to be transmitted, and apply the terminal subchannel identifier to each Chirp symbol using a frequency wrap-around cyclic shift method. Generate and send a feature signal carrying a frequency offset fingerprint; The signal receiving module, located at the gateway, is used to continuously monitor the uplink signals of all terminals and analyze the mixed conflict signals by delinear frequency modulation and fast Fourier transform. The signal separation and decoding module, located in the gateway, is used to analyze each peak in the Fast Fourier Transform spectrum and calculate the signal strength of each peak. According to different The peak signal is separated into the corresponding sub-channel stream, and each sub-channel stream is decoded in parallel to obtain the raw uplink data from different terminals.

[0025] Furthermore, it also includes a dynamic scheduling module, located in the gateway, which is used to dynamically adjust the Δf value allocated to the terminal through the downlink based on network load and conflict history, so as to realize the adaptive allocation and scheduling of sub-channel resources.

[0026] Another object of the present invention is to provide an electronic device, including a processor and a memory storing a computer program, characterized in that the processor, when executing the computer program, implements the LoRa uplink channel capacity extension method based on controllable frequency offset provided by the first object of the present invention.

[0027] Another object of the present invention is to provide a computer-readable storage medium having a computer program stored thereon, characterized in that, when the computer program is executed by a processor, it implements the LoRa uplink channel capacity extension method based on controllable frequency offset provided by the first object of the present invention.

[0028] In combination with the above technical solutions, the beneficial effects of the present invention compared with the prior art are as follows: Channel capacity is doubled by introducing controllable frequency offset. As a new quasi-orthogonal dimension, a physical channel is virtually divided into multiple logical sub-channels, which can multiply the uplink channel logical capacity without increasing physical hardware resources. Collision resolution capability: The gateway gains the ability to separate and decode multiple data packets from time-frequency overlapping collision signals, and can proactively identify them. Fingerprint recognition has changed the situation of "collision equals packet loss" in LoRa networks; Active controllability and high compatibility: It is actively configured and scheduled by the network side, which is highly flexible; the signal generated by the "cyclic shift" method is completely within the standard channel bandwidth and is compatible with the spectrum of existing commercial LoRa devices. Support for heterogeneous terminal convergence and unified management: This invention provides a unified capacity expansion solution covering two types of terminals: current mainstream (active radio frequency terminals) and future ultra-low power (backscatter terminals). It solves the key problem of lack of standardized access and capacity expansion methods when emerging backscatter terminals coexist with traditional networks, and removes technical obstacles for the large-scale, heterogeneous convergence deployment of the Internet of Things.

[0029] This invention directly reduces the most energy-consuming radio frequency transmission activities of terminals by reducing collisions and retransmissions. For battery-powered IoT terminals, this will significantly extend their service life and reduce maintenance costs, which has vital economic value for large-scale IoT applications. Attached Figure Description

[0030] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart of the LoRa uplink channel capacity expansion method based on controllable frequency offset provided in the embodiments of the present invention; Figure 2 This is a schematic diagram illustrating the effect of signal-to-noise ratio on demodulation provided in an embodiment of the present invention; Figure 3 This is a statistical diagram illustrating the number of transmissions required for each data packet according to an embodiment of the present invention; Figure 4 This is a schematic diagram showing the number of terminals and the average number of uplink transmissions provided in the embodiments of the present invention; Figure 5 In the diagram, (a) represents the average transmission time of the data packet when the symbol is repeated a different number of times. Figure 5 (b) in the figure represents the average number of uploads per data packet; Figure 6 This is a schematic diagram of single-channel throughput provided in an embodiment of the present invention; Figure 7 This is a schematic diagram of single-channel delay provided in an embodiment of the present invention; Figure 8 This is a schematic diagram of the LoRa uplink channel capacity expansion system based on controllable frequency offset provided in an embodiment of the present invention. Detailed Implementation

[0031] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0032] Example 1:

[0033] like Figure 1 The image shows an embodiment of the LoRa uplink channel capacity extension method based on controllable frequency offset provided by the present invention, which includes the following steps: S1: The gateway assigns a unique sub-channel identifier to each terminal that needs uplink transmission. The sub-channel identifier is associated with a predefined controllable frequency offset value. ; S2: The terminal generates standard LoRa symbols based on the data to be transmitted, and applies the terminal's sub-channel identifier to each Chirp symbol using a frequency-wrap cyclic shift method. Generate and send a feature signal carrying a frequency offset fingerprint; S3: The gateway continuously monitors the uplink signals of all terminals and analyzes the mixed conflict signals by delinear frequency modulation and fast Fourier transform. S4: The gateway analyzes each peak in the Fast Fourier Transform spectrum and calculates the signal of each peak. According to different The peak signal is separated into the corresponding sub-channel stream, and each sub-channel stream is decoded in parallel to obtain the raw uplink data from different terminals.

[0034] Specifically, this embodiment provides a LoRa uplink channel capacity expansion method and system based on controllable frequency offset (Δf). Its core lies in introducing a "controllable, small frequency offset (Δf)" as a new, quasi-orthogonal signal dimension into the LoRa physical layer, thereby virtually dividing a standard physical uplink channel into multiple logical "sub-channels." By recognizing this "fingerprint," the gateway can separate and decode data packets from different terminals from time-frequency overlapping conflicting signals.

[0035] To overcome the bottleneck of "limited number of physical channels," this invention proposes to use controllable frequency offset (Δf) as a new, quasi-orthogonal signal dimension. By virtually dividing a physical channel into multiple logical sub-channels with different frequency offset (Δf) "fingerprints," the logical capacity of the uplink channel is significantly improved without increasing the number of physical radio frequency channels, thus achieving "open source."

[0036] To enable gateways to handle collisions in high-density deployments, this invention aims to allow gateways to proactively identify and utilize the "controllable frequency offset (Δf)" feature in terminal signals. Upon receiving time-frequency overlapping collision signals, the gateway can separate the signal components into different sub-channel streams based on different frequency offset (Δf) amounts, and perform parallel decoding, thereby transforming "collision equals failure" into "collision is parseable".

[0037] To overcome the passivity and inflexibility of the "using inherent hardware frequency offset (Δf)" approach, this invention introduces a frequency offset (Δf) that is actively generated, network-configurable, and dynamically schedulable. The gateway can allocate or adjust the frequency offset (Δf) value of the terminal via the downlink based on network load, thereby achieving optimized resource allocation.

[0038] To accommodate the heterogeneous trend of IoT terminals, this invention provides a unified solution covering both active radio frequency (RF) terminals and backscatter terminals. In particular, it designs a standard-compliant channel access and capacity expansion mechanism for LoRa backscatter terminals, solving the challenges of coexistence with traditional networks and large-scale deployment.

[0039] Through the aforementioned capacity expansion and collision resolution, the ultimate goal is to significantly reduce the probability of packet collisions and the number of retransmissions, thereby increasing network throughput and reducing end-to-end latency. Simultaneously, it reduces the ineffective transmission power consumption of terminals, significantly extending the operational lifespan of battery-powered IoT terminals and lowering maintenance costs.

[0040] This invention introduces a proactive and flexible new dimension of "controllable frequency offset (Δf)" to virtually expand channel capacity at the physical layer, enabling gateways to separate conflicting signals. This systematically solves a series of performance bottlenecks and compatibility issues caused by existing technologies' inability to overcome physical channel limitations and reliance on uncontrollable or non-ideal signal characteristics, ultimately enabling reliable, efficient, and large-scale deployment of LoRa networks in high-density scenarios.

[0041] The entire solution constitutes a complete communication chain, involving gateways, terminals, and their collaborative work. The main process can be summarized into four core steps: Step 1: Sub-channel mapping and configuration Executor: Gateway.

[0042] Action: The gateway assigns a unique sub-channel identifier to the terminal that needs uplink transmission. This identifier is associated with a predefined, small frequency offset value Δf.

[0043] Key constraint: Δf must satisfy Where BW is the channel bandwidth and SF is the spreading factor. This constraint ensures that the frequency offset (Δf) does not interfere with the data encoding information carried by the symbol itself. Configuration method: This configuration is completed through the downlink. For backscatter terminals, the Δf information can be implicitly indicated by the encoded value of the downlink LoRa excitation signal itself.

[0044] Step 2: Characteristic Signal Generation and Transmission Executor: Terminal.

[0045] Action: After the terminal generates a standard LoRa symbol based on the data to be transmitted, it applies it to each Chirp symbol in a "cyclic shift" manner with frequency wrapping, according to the Δf assigned to it.

[0046] Detailed Explanation of "Cyclic Shift": The starting frequency of a symbol is increased by Δf based on the standard coding frequency, and its frequency changes linearly with time. The key point is that when the instantaneous frequency reaches the upper limit of the channel bandwidth, it does not continue to increase linearly beyond the bandwidth, but instead wraps around to the lower limit of the bandwidth and starts increasing linearly again. This ensures that the energy of the entire symbol is always constrained within the standard channel bandwidth, thereby maintaining physical layer spectrum compatibility with standard LoRa devices.

[0047] The specific logic principle of frequency wrap-around is as follows: during the linear modulation of the LoRa Chirp symbol frequency, its instantaneous frequency is forced to jump when it reaches the channel bandwidth boundary and return to the opposite boundary, so as to ensure that the energy of the entire symbol is constrained within a fixed and continuous frequency range (i.e., the standard channel bandwidth) from beginning to end.

[0048] In this invention, "cyclic shifting" is the means, and "frequency wrap-around" is the underlying physical layer rule upon which this means depends. By applying Δf to the starting frequency that follows the wrap-around rule, physical layer compatibility is achieved, ensuring that the generated signal fully conforms to the standard LoRa spectral template and can be received by any standard LoRa receiver (although only a custom gateway can interpret Δf). It also concentrates all signal energy within a predetermined bandwidth, avoiding energy dispersion and out-of-band radiation caused by simple mixing. Different Δf values, under the same wrap-around rule, result in signals that are phase-shifted on the same "cyclic frequency axis" in the time-frequency domain, thus providing a theoretical basis for the gateway to separate and identify conflicting signals.

[0049] Output: The terminal sends this characteristic signal along with the "frequency offset (Δf) fingerprint".

[0050] Step 3: Collision Signal Reception and Feature Extraction Executor: Gateway.

[0051] Action: The gateway receives uplink signals (potentially mixed signals from collisions) from all terminals within its antenna range and performs standard de-chirp and Fast Fourier Transform (FFT) operations. Characteristics: In an ideal single-user scenario, the energy of a chirp symbol will concentrate at a single frequency point (peak). However, in a multi-user collision scenario, multiple peaks will appear in the FFT spectrum, each corresponding to a component of a colliding chirp symbol.

[0052] Step 4: Signal separation and parallel decoding based on frequency offset (Δf) Executor: Gateway.

[0053] action: 1. Frequency offset (Δf) identification: The gateway analyzes each peak in the FFT spectrum and calculates the minimum difference between its frequency and the theoretical frequency of all possible standard symbol codes. This difference is the Δf of the signal component.

[0054] 2. Signal Separation: Based on the identified different Δf values, these peaks are classified into different sub-channel streams. Each sub-channel stream uniquely corresponds to a specific source terminal.

[0055] 3. Parallel Decoding: The gateway performs independent, standard LoRa decoding on the peak sequence within each sub-channel stream, thereby recovering the original uplink data from different terminals in parallel.

[0056] Dynamic optimization: The gateway can dynamically adjust the Δf value allocated to the terminal through the downlink based on network load and conflict history, thereby achieving adaptive allocation and scheduling of sub-channel resources and further improving network efficiency.

[0057] The gateway determines network load not directly based on the number of terminals, but rather on the conflict status of uplink data packets. Specifically: Collision Detection: During demodulation, the gateway may detect multiple energy peaks in the FFT spectrum within the same demodulation window, and the frequencies corresponding to these peaks cannot be interpreted using a single, standard symbol encoding value (i.e., Δf exists). When there are too many of these peaks and they cannot be effectively separated, it indicates a severe collision.

[0058] Conflict history record: The gateway can count the frequency of decoding failures or retransmissions on each sub-channel (Δf value) as a quantitative indicator of whether the sub-channel is "congested" or "high-loaded".

[0059] Depending on the different load conditions, the gateway will adopt different Δf adjustment strategies: Low load / Initial allocation state: When the network starts up or a terminal joins the network, the load is relatively light. The gateway adopts a static or simple round-robin allocation strategy. That is, a newly joined terminal is assigned a currently unoccupied and distinct Δf value, ensuring that each terminal works on a different logical sub-channel in the initial state, avoiding conflicts caused by random access.

[0060] When a device joins the network, the gateway cyclically allocates sub-channels to all terminals. This has the advantage of eliminating the waste of channel resources caused by random channel selection when the number of devices is small.

[0061] Medium-to-high load / collision status: When the gateway detects increased collisions on a specific sub-channel (e.g., Δf0), it indicates that the "virtual channel" is overloaded. The gateway's scheduling and management unit will proactively intervene, instructing one or more terminals transmitting data on the conflicting channel via downlink (e.g., by carrying instructions in the ACK frame) to switch their assigned Δf values ​​to other Δf values ​​that are currently idle or have fewer collisions. This "disperses" the conflicting terminals to different logical sub-channels, achieving load balancing. The gateway can dynamically schedule sub-channel allocation, adjusting the Δf values ​​of terminals via downlink based on network load and collision history, achieving adaptive resource allocation.

[0062] Under conditions of extremely high load / resource scarcity: When the number of terminals approaches or exceeds the total number of available Δf values ​​(i.e., the upper limit of logical sub-channel capacity), simple adjustments cannot avoid conflicts. In this situation, dynamic scheduling may require a combination of more complex frequency hopping mechanisms or transmission timing control. For example, a gateway can instruct a group of terminals to use the same Δf but with different frequency hopping sequences, avoiding continuous conflicts through time staggering. Alternatively, the gateway can schedule non-urgent terminals to delay transmission, prioritizing the transmission of important data. This maximizes the number of successfully decoded data packets under resource constraints, thereby improving network throughput.

[0063] The dynamic scheduling of a gateway is a complete loop that includes monitoring, decision-making, execution, and feedback. The specific steps are as follows: (1) Initial allocation and configuration (terminal network access / network initialization) Action: When a new terminal device needs to join the network or when the network initializes, the gateway's "scheduling management unit" assigns a unique "sub-channel identifier" to each terminal. This identifier is associated with a specific, small frequency offset Δf. This configuration information is sent to the terminal via the downlink.

[0064] Implementation method: For active radio terminals: configuration instructions may be encapsulated in LoRaWAN MAC commands. After the terminal receives the Δf, its software-defined radio (SDR) module will use it when generating subsequent signals.

[0065] For backscatter terminals: The gateway transmits a specifically coded LoRa excitation signal downlink, the coded value of which implicitly contains subchannel identification information. The low-power demodulation circuit on the terminal demodulates this code and configures the local voltage-controlled oscillator (VCXO) accordingly to generate the corresponding subcarrier frequency, thereby introducing a specified Δf into the reflected signal.

[0066] (2) Uplink conflict monitoring and load assessment Action: The gateway continuously receives uplink signals from all terminals. When multiple terminal signals overlap (collide) in the time-frequency domain, the gateway's "signal processing and frequency offset identification unit" performs de-chirp and FFT operations.

[0067] Evaluation basis: Gateway analysis of FFT spectrum: Collision detection: Multiple energy peaks are identified, and the frequencies corresponding to these peaks have slight differences (i.e., different Δf) from the theoretical frequencies of the standard symbol encoding.

[0068] Identifying load hotspots: The gateway records and analyzes which Δf values ​​(i.e., which sub-channels) frequently have multiple peaks that cannot be effectively separated (decoding failure), or which Δf values ​​correspond to frequent packet retransmission requests. This constitutes the "network load and conflict history".

[0069] (3) Scheduling decision and instruction generation Action: The gateway's "scheduling management unit" makes dynamic scheduling decisions based on the monitoring data from step 2.

[0070] Decision-making logic: Load balancing: If the gateway finds that the collision rate of a sub-channel corresponding to a certain Δf is significantly higher than that of other sub-channels, it will decide to adjust some terminals on that sub-channel to other Δf values ​​that are currently less busy.

[0071] Resource optimization: The gateway may dynamically adjust the available Δf set based on the number of active terminals and data transmission patterns in the network (e.g., when the network is extremely congested, smaller Δf steps may be enabled to create more sub-channels, but this may reduce the demodulation robustness of individual sub-channels).

[0072] Generate instructions: The scheduling management unit generates downlink configuration instructions containing information such as "target terminal address" and "new sub-channel identifier (new Δf value)".

[0073] (4) Downlink command transmission and terminal reconfiguration Action: The gateway sends the generated scheduling instructions to the specific target terminal via the downlink.

[0074] Terminal execution: After receiving the instruction, the terminal device: Active RF terminal: Its processing unit (such as MCU) parses the instructions and reconfigures the signal generation parameters of the SDR module. Subsequent uplink signals will use the new Δf.

[0075] Backscatter terminal: Its low-power demodulation circuit demodulates commands and adjusts the output voltage of the DAC, thereby changing the oscillation frequency of the VCXO and generating a new subcarrier. The uplink signal generated by the subsequent reflection carries a new Δf "fingerprint".

[0076] Action complete: The terminal has completed the local Δf value update.

[0077] (5) Effect verification and closed loop Action: The terminal sends uplink data using the new Δf value. The gateway receives and demodulates the data again, monitoring whether the collision situation in this sub-channel has been alleviated after the adjustment.

[0078] Closed loop: Based on the new monitoring data, the gateway can decide whether further adjustments are needed, thus forming an adaptive closed-loop control system of "monitoring → decision-making → execution → re-monitoring".

[0079] It should be noted that in this embodiment, the three main characteristics of "active generation, network configurability, and dynamic scheduling" are integrated throughout the above steps and the entire system: (1) Active generation This demonstrates that the frequency offset Δf is not an inherent, uncontrollable defect of the terminal hardware (such as inherent frequency offset between chips), but rather a "feature" or "fingerprint" actively applied to the signal by the terminal.

[0080] Implementation method: On the active RF terminal, cyclically shifted Chirp symbols are directly generated during the baseband signal processing stage via SDR to precisely control Δf.

[0081] At the backscatter terminal, a specific subcarrier is generated by controlling the local VCXO and actively selecting the path to mix with the excitation signal, thereby generating a reflected signal carrying a specified Δf.

[0082] Difference from existing technologies: This differs from solutions that utilize "inherent hardware frequency offset," which rely on uncontrollable dimensions and have poor flexibility.

[0083] (2) Network configurable This means that the Δf value used by the terminal is not pre-programmed or randomly selected, but is remotely configured and managed by the network side (gateway) through the downlink.

[0084] Implementation method: Whether it's the initial allocation or subsequent adjustments, the value of Δf originates from the instructions issued by the gateway.

[0085] This provides a unified control plane for network management, enabling gateways to grasp and manage the uplink channel "fingerprints" of all terminals, laying the foundation for efficient scheduling and multiple access.

[0086] (3) Dynamic scheduling is possible This means that the Δf value used by the terminal is not fixed. The gateway can actively and dynamically adjust the Δf allocation of one or more terminals based on real-time network load status, conflict history, and other information.

[0087] Implementation method: The gateway has a "scheduling management unit" that can make decisions based on monitoring data and update the terminal configuration in real time through the downlink.

[0088] This allows network resources (virtual subchannels) to be centrally and intelligently allocated and reused, much like frequency resources in cellular networks, thereby responding to changes in network load, maximizing overall throughput, and reducing collisions and retransmissions.

[0089] This embodiment fully realizes the "active, configurable, and schedulable" nature of frequency offset through a technical chain of "terminal actively generating characteristic signals + gateway centrally configuring and managing characteristic parameters" + "dynamically adjusting parameters based on load feedback". This fundamentally transforms LoRa uplink access from a completely random and passive ALOHA mechanism to a partially controllable and optimizable new paradigm, significantly improving the scalability and performance of the network in high-density deployments.

[0090] The technical effects of the present invention will be further explained below with reference to specific experiments.

[0091] 1. SDR Implementation The transceivers for the terminal and gateway are implemented based on the USRP2920 software-defined radio device, using GNU Radio as the software platform. GNU Radio is an open-source software development kit specifically designed for building software-defined radio (SDR) systems. It provides a rich set of signal processing modules, which users can flexibly combine to implement complex wireless communication systems. In practical applications, the USRP (Universal Software-Defined Radio Peripheral) primarily serves as the data interface, responsible for transmitting and receiving wireless signals. The combination of this hardware and the GNU Radio software platform allows researchers and developers to directly control the signal processing flow through software, providing high flexibility and controllability.

[0092] In the terminal section, to generate compliant LoRa symbols, the content to be transmitted is first processed accordingly. The transmitter performs a slight cyclic frequency shift based on the standard LoRa symbol, and we implemented a module that can generate symbols based on a specified frequency offset and symbol value. The generated symbols are transmitted via the USRP, and up-conversion is performed simultaneously.

[0093] At the gateway receiver, the received signal is first de-chirped, then FFT transformed to extract all peak values. Based on the offset of the peak value relative to the original LoRa symbol encoding, the signal is classified into different ports, i.e., divided into different data packets, and then a series of decoding operations are performed.

[0094] 2. Recognition accuracy test To test the success rate of frequency offset identification under different signal-to-noise ratios (SNRs), a USRP was used as a terminal to transmit LoRa symbols with various frequency offsets under different signal-to-noise ratios (SF). Another USRP-implemented gateway received and stored the data, then processed it using MATLAB, and the success rate of identifying the additional frequency offset of LoRa symbols was statistically analyzed. Random chirp signals of the same frequency were introduced as interference in the experiment. The experiment showed that the additional frequency offset could be identified by increasing the number of FFT points, but this increase was not unlimited. To ensure identification effectiveness, the minimum frequency offset was set to Δ. At this point, the number of sub-channels corresponding to one uplink channel was 4. The results are as follows: Figure 2 As shown, at a low SNR (Signal-to-noise Ratio), the frequency peak after demodulation is distorted by the influence of adjacent interference frequency components. Since the FFT result at the gateway only searches for the peak, it will cause the identification error of frequency offset. If the additional frequency offset is to be further improved, a more complex algorithm can be used for identification.

[0095] 3. Uplink transmission count experiment (a) Experiment on average upload count under congested channel To test the impact of introducing subchannels on system performance, uplink collision scenarios of LoRa terminals were simulated using MATLAB. The experiment followed these conditions: the terminal operated in CLASS A mode, employing a binary exponential backoff algorithm; and uplink transmission was performed using the defined subchannels.

[0096] To test the uplink transmission performance of this system, a typical LoRa uplink collision scenario was considered. When a LoRa system experiences a large-scale uplink failure due to unforeseen circumstances, each terminal randomly begins uploading within 10 seconds. Once an upload is successful, the terminal stops uploading. If a collision occurs, the system re-initiates a retransmission after a random backoff until a successful upload is achieved. To test its performance at different SNR levels, experiments were conducted at three different SNR levels: low SNR < -10dB, medium SNR < -5dB, and high SNR < 5dB, respectively. The experimental parameters were set as follows: 20 terminals, a spreading factor of 7, and a fixed 27-byte payload data with a payload length of 30 symbols.

[0097] Ultimately, different levels of effectiveness were observed with varying degrees of added frequency offset. Traditional LoRa terminals experienced uplink congestion, leading to numerous retransmissions until the retransmission limit was reached. However, after introducing "frequency offset" to expand the uplink channels, the uplink congestion was alleviated. With two sub-channels, the number of uplink failures was significantly reduced, decreasing the number of upload attempts by approximately 60%; with four sub-channels, the number of uplink failures was almost zero. Figure 3 As shown, the number of retransmissions by the terminal directly affects the lifespan of the entire system, and the number of uplink failures affects the reliability of system data. By "expanding" the uplink channel, uplink conflicts in LoRa IoT have been mitigated.

[0098] Furthermore, frequency shift recognition can be affected by environmental noise. Especially at low signal-to-noise ratios (SNR), frequency peaks can be distorted to some extent by environmental noise, leading to recognition failure. Under medium to high SNR conditions, additional frequency shifts can be extracted more accurately, thus distinguishing different symbols. At low SNR, frequency peak distortion causes recognition errors, preventing the gateway from demodulating properly.

[0099] Of course, in practical applications, in order to achieve longer transmission distances or data rate requirements, terminal devices need to be set with different spreading factors, which will also change the air time of transmission and thus affect the probability of collisions.

[0100] (II) Small-scale terminal deployment experiment Under high signal-to-noise ratio conditions, the number of connected terminal devices also affects the number of upload attempts required for the terminal to upload data. Gradually increasing the number of devices in the system, even when the number of devices is small, will lead to an increased probability of uplink collisions for LoRa terminals based on the Aloha mechanism. Figure 4 As shown, when gateway control is used to schedule the uplink channel of the terminal device, the lifespan of battery-powered terminals can be greatly improved because it alleviates the conflicts caused by random channel selection.

[0101] (III) Experiment on sending data copies Retransmitting a symbol n times, while keeping the sampling frequency constant, can increase the sampling time by n times for the same symbol, thereby further refining the frequency offset that can be added and increasing the number of sub-channels. However, sending data copies also increases the data transmission time. Figure 5As shown in (a) and (b), this embodiment tested the uplink transmission of 40 data packets randomly within 10 seconds with SF=7 and a symbol length of 8. The average transmission time was then calculated for different numbers of replicas. When the number of replicas was 1, 2, and 3, the corresponding number of sub-channels was 4, 8, and 16. Because the number of sub-channels increased, reducing the number of uplink transmissions, although the transmission time for each data packet increased, the average transmission time did not increase linearly with the number of replicas sent, despite reducing uplink data packet transmission conflicts. When the number of replicas was 4, the corresponding number of sub-channels was 32. The effect of increasing the number of sub-channels on reducing uplink data packet transmission conflicts became marginal, and the average transmission time increased accordingly. This indicates that in practical applications, the number of data replicas can be increased by sending a certain number of data replicas based on the upload activity of IoT terminals in the network.

[0102] (iv) Reduce the number of symbol codes and increase the number of sub-channels To increase the number of sub-channels, the data rate can be reduced. For example, when the spreading factor is 7, reducing the number of encoded bits by 1 can correspondingly increase the number of sub-channels by a factor of 2. The LoRa data rate calculation formula is DR = SF × (BW / 2). SF )×CR, where CR is the coding rate, which takes values ​​from 1 to 4. The number of sub-channels corresponding to the data rate when the bandwidth is 125kHz, SF=7, and CR=1 is shown in Table 1.

[0103] Table 1. Reducing the data rate and increasing the number of sub-channels 4. Network throughput experiment Uplink network throughput is a crucial metric, representing the scale of terminal data received by the gateway per unit time. In the test, only one LoRa uplink channel was used, with a spreading factor of 7 and a bandwidth of 125kHz. Subchannels were allocated cyclically. The number of devices was gradually increased, and throughput under different conditions was statistically analyzed. To increase the number of subchannels, the number of bits that can be encoded per symbol needs to be reduced accordingly. When the number of subchannels is 8, one chirp symbol can transmit 6 bits; when the number of subchannels is 16, one chirp symbol can transmit 5 bits. The throughput is as follows: Figure 6 As shown, when the number of devices is less than the number of sub-channels, each device can be allocated an independent sub-channel. When the number of devices exceeds the number of sub-channels, data conflicts will occur, leading to a decrease in throughput. To further increase the number of devices, a relatively low data rate can be adopted.

[0104] Furthermore, this work was compared with other works, and the results showed that its network throughput was superior to choir. This is because choir uses an inherent hardware decimal point frequency offset to distinguish data packets, which is uncontrollable and essentially means that its sub-channel selection is random. In contrast, this work avoids resource waste caused by conflicts between two terminals due to random sub-channel selection by cyclically allocating sub-channels.

[0105] Furthermore, increasing the number of sub-channels by reducing the data rate is feasible. Although this will increase the uplink time for each terminal, it can improve the overall network throughput. If other channels with different carrier frequencies are enabled simultaneously, the network throughput will increase proportionally, supporting more devices.

[0106] 5. Uplink Delay Experiment In LoRa IoT systems, uplink latency is also a crucial metric. Here, uplink latency is defined as the time it takes for the terminal device to send uplink data to the gateway, and for the gateway to correctly receive that data. In the test, only one LoRa uplink channel was used, with a spreading factor of 7, a bandwidth of 125kHz, and a cyclic allocation method for the sub-channels. The experimental results are as follows: Figure 7 As shown, LoRa terminals using only Aloha channel access experience increased retransmission backoff time and thus increased average uplink latency as their number increases. This work reduces uplink latency by allocating different sub-channels to each terminal as much as possible, thereby reducing retransmission backoff time. The uplink latency of this work is higher than that of Choir because increasing the number of sub-channels reduces the data rate, leading to longer uplink data transmission times.

[0107] Conclusion: The experiment implemented a software-defined radio model for a LoRa terminal based on GNU Radio, which satisfies the LoRa physical layer protocol and allows for frequency offset addition. By testing the average number of retransmissions per data upload under various conditions, the results show that under congested uplink channel conditions, introducing a subchannel mechanism can reduce the number of retransmissions by approximately 60%, which is crucial for LoRa terminal devices where the number of uploads determines their operational lifespan. Similarly, we tested the uplink network throughput, and the results suggest that we can increase the number of subchannels by reducing the data rate, which can maintain a certain network throughput even with a large number of devices, while also significantly reducing uplink latency.

[0108] Example 2:

[0109] like Figure 8As shown, this embodiment provides a LoRa uplink channel capacity extension system based on controllable frequency offset, including: The configuration module, located on the gateway, is used to assign a unique sub-channel identifier to each terminal requiring uplink transmission. Each sub-channel identifier is associated with a predefined, controllable frequency offset value. ; The signal generation module, located in the terminal, is used to generate standard LoRa symbols based on the data to be transmitted, and apply the terminal subchannel identifier to each Chirp symbol using a frequency wrap-around cyclic shift method. Generate and send a feature signal carrying a frequency offset fingerprint; The signal receiving module, located at the gateway, is used to continuously monitor the uplink signals of all terminals and analyze the mixed conflict signals by delinear frequency modulation and fast Fourier transform. The signal separation and decoding module, located in the gateway, is used to analyze each peak in the Fast Fourier Transform spectrum and calculate the signal strength of each peak. According to different The peak signal is separated into the corresponding sub-channel stream, and each sub-channel stream is decoded in parallel to obtain the raw uplink data from different terminals.

[0110] Preferably, this embodiment also includes a dynamic scheduling module, which is set in the gateway and is used to dynamically adjust the Δf value allocated to the terminal through the downlink according to the network load and conflict history, so as to realize the adaptive allocation and scheduling of sub-channel resources.

[0111] Example 3:

[0112] This embodiment provides an electronic device, including a processor and a memory storing a computer program, characterized in that the processor executes the computer program to implement the LoRa uplink channel capacity expansion method based on controllable frequency offset provided in Embodiment 1 of the present invention.

[0113] Example 4:

[0114] This embodiment provides a computer-readable storage medium storing a computer program, characterized in that the computer program, when executed by a processor, implements the LoRa uplink channel capacity expansion method based on controllable frequency offset provided in Embodiment 1 of the present invention.

[0115] It should be understood that although the steps in the flowcharts of the various embodiments of the present invention are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated in the present invention, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least some steps in the various embodiments may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a portion of the sub-steps or stages of other steps.

[0116] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided by this invention can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0117] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for extending the uplink channel capacity of LoRa based on controllable frequency offset, characterized in that, The method includes the following steps: The gateway allocates a unique subchannel identity to terminals requiring uplink transmission, the subchannel identity being associated with a predefined controllable frequency offset value ; The terminal generates standard LoRa symbols based on the data to be transmitted, and applies the terminal subchannel identifier corresponding to each Chirp symbol in a frequency-wrap cyclic shift manner. Generate and send a feature signal carrying a frequency offset fingerprint; The gateway continuously monitors the uplink signals of all terminals and analyzes the mixed conflict signals by delinear frequency modulation and fast Fourier transform. The gateway analyzes each peak in the Fast Fourier Transform spectrum and calculates the signal of each peak. According to different The peak signal is separated into the corresponding sub-channel stream, and each sub-channel stream is decoded in parallel to obtain the raw uplink data from different terminals.

2. The method of claim 1, wherein, The frequency offset value Δf satisfies a constraint condition wherein is the channel bandwidth, is the spreading factor. 3.The LoRa uplink channel capacity expansion and system based on controllable frequency offset of claim 1, wherein, The cyclic shift method specifically involves: forcing the instantaneous frequency of the Chirp symbol to jump and wrap back to the opposite boundary when it reaches the channel bandwidth boundary during linear modulation, ensuring that the symbol energy is always constrained within the standard channel bandwidth.

4. The method of claim 1, wherein, The terminal includes an active radio frequency terminal and a backscatter terminal; For active radio frequency terminals, software-defined radio is used to directly generate cyclically shifted Chirp symbols during the baseband signal processing stage; For backscatter terminals, Δf is obtained by demodulating the downlink excitation signal command of the gateway, and a local voltage-controlled oscillator is configured to generate a subcarrier with a frequency corresponding to the subchannel. After mixing and reflection, a reflected signal is generated that is located in the standard uplink channel and carries the characteristics of the specific subchannel.

5. The method of claim 1, wherein, The gateway dynamically adjusts the Δf value allocated to the terminal via the downlink based on network load and conflict history, achieving adaptive allocation and scheduling of sub-channel resources, specifically including: Monitoring steps: The gateway continuously receives uplink signals, demodulates and identifies each element in the Fast Fourier Transform spectrum. Record the frequency of decoding failure or retransmission corresponding to the value, and record the collision rate of each sub-channel. Decision-making steps: The scheduling management unit analyzes conflict data, identifies load hotspots, and determines the terminals and target Δf values ​​that need to be adjusted; Execution steps: Send a reconfiguration command to the target terminal via the downlink; Feedback steps: The target terminal updates its local Δf value, and subsequent uplink data is sent using the new Δf value.

6. The method of claim 5, wherein, The gateway employs different Δf value adjustment strategies based on different load conditions, including: Initial allocation state: The gateway adopts a round-robin allocation strategy to allocate a unique, unused Δf to the terminal, eliminating random access conflicts; Conflict occurrence status: The gateway switches the terminals on the high-conflict sub-channel to the currently idle or less conflicted sub-channel to achieve load balancing; In resource-scarce situations: The gateway combines frequency hopping mechanisms and transmission timing control to prioritize the transmission of important data and maximize the decoding success rate.

7. A system for LoRa uplink channel capacity expansion based on controllable frequency offset, characterized in that, The system includes: The configuration module is arranged in the gateway and is configured to assign a unique subchannel identifier to each terminal that needs to perform uplink transmission, wherein the subchannel identifier is associated with a predefined controllable frequency offset value ; The signal generation module, located in the terminal, is used to generate standard LoRa symbols based on the data to be transmitted, and apply the terminal subchannel identifier to each Chirp symbol using a frequency wrap-around cyclic shift method. Generate and send a feature signal carrying a frequency offset fingerprint; The signal receiving module, located at the gateway, is used to continuously monitor the uplink signals of all terminals and analyze the mixed conflict signals by delinear frequency modulation and fast Fourier transform. The signal separation and decoding module, located in the gateway, is used to analyze each peak in the Fast Fourier Transform spectrum and calculate the signal strength of each peak. According to different The peak signal is separated into the corresponding sub-channel stream, and each sub-channel stream is decoded in parallel to obtain the raw uplink data from different terminals.

8. The system for LoRa uplink channel capacity expansion based on controllable frequency offset according to claim 7, characterized in that, It also includes a dynamic scheduling module, located in the gateway, which is used to dynamically adjust the Δf value allocated to the terminal through the downlink based on network load and conflict history, so as to realize the adaptive allocation and scheduling of sub-channel resources.

9. An electronic device comprising a processor and a memory having a computer program stored therein, characterized in that, When the processor executes the computer program, it implements the LoRa uplink channel capacity expansion method based on controllable frequency offset as described in any one of claims 1 to 6.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, When the computer program is executed by the processor, it implements the LoRa uplink channel capacity expansion method based on controllable frequency offset as described in any one of claims 1 to 6.

Images