Frequency hopping sequence generation method based on channel quality in channel heterogeneous scene

Abstract

The invention provides a frequency hopping sequence generation method based on channel quality in a channel heterogeneous scene. The method comprises the following implementation steps: initializing a self-organizing network; calculating the quality of each global channel and constructing a guide sequence; calculating the reference length of a sub-sequence in the frequency hopping sequence to be generated; constructing a lambda-type frequency hopping subsequence based on channel quality; constructing an R-type frequency hopping subsequence through the guide sequence; and obtaining a frequency hopping sequence generation result. According to the invention, the rows in the frequency hopping matrix constructed by the four lambda frequency hopping subsequences with coprime lengths generated according to the guide sequence and the R-type frequency hopping subsequences are sequentially spliced to obtain the frequency hopping sequence, so that the nodes can successfully realize channel convergence within finite time; and meanwhile, the normalized signal to interference plus noise ratio of each global channel is used as the quality of the global channel, and the available channel set is mapped based on the channel quality to construct the lambda-type frequency hopping subsequence, so that the influence of heterogeneous interference on channel convergence is reduced, and the reliability of node convergence is improved.

Description

Technical Field

[0001] The present invention belongs to the field of communication technologies, and relates to a method for generating a frequency-hopping sequence. Specifically, it relates to a method for generating a frequency-hopping sequence based on channel quality in a channel heterogeneous scenario, and can be used in self-organizing networks with asynchronous clocks, heterogeneous channels, and heterogeneous interferences. Background Art

[0002] A self-organizing network is a wireless network system that does not require the support of fixed infrastructure and realizes distributed communication through autonomous organization of nodes. In a channel heterogeneous scenario, due to the dynamic changes of local spectra, the available channel sets between devices are often different. To establish an effective communication channel, two or more nodes must simultaneously access the common available channels and exchange control information, and this process is called rendezvous. Traditional rendezvous methods based on a common control channel greatly limit their applicability in scenarios with complex and dynamic environments because they require reserved channel resources and are vulnerable to external interferences and malicious attacks. In contrast, the blind rendezvous frequency-hopping method does not rely on a fixed common control channel, but generates a set of frequency-hopping sequences by all nodes in the network following a unified rule. Different nodes can achieve rendezvous on different channels based on the frequency-hopping sequences in the same time slot. The method for generating frequency-hopping sequences is a key factor affecting the establishment of efficient and reliable communication channels between nodes in a self-organizing network based on frequency-hopping rendezvous.

[0003] To improve the rendezvous efficiency in heterogeneous channels, researchers have proposed many solutions. For example, in a patent document with the application publication number CN114726402A and the title "An Anonymous Frequency-Hopping Sequence Design Method for Multi-Antenna Cognitive Wireless Networks", a method for generating a frequency-hopping sequence is disclosed, which includes numbering multiple antennas of cognitive nodes and dividing the available channels into multiple channel groups; initializing a frequency-hopping sequence for each antenna and generating a frequency-hopping sequence based on 3 channels frame by frame; generating a periodic frequency-hopping sequence of 15 time slots through binary and undecimal coding techniques, combined with the combination of the first type and the second type of disjoint difference sets, and concatenating and expanding it into a complete frequency-hopping sequence. This invention improves the frequency-hopping rendezvous ability and the interaction performance of control information, but it depends on fixed channel group division and has insufficient adaptability in a channel heterogeneous scenario and does not consider the impact of heterogeneous interference on channel rendezvous, which may cause nodes to frequently meet on high-interference channels, increasing the packet loss and retransmission overhead in the communication process, thus affecting the channel rendezvous efficiency and the reliability of communication. Summary of the Invention

[0004] The purpose of the present invention is to propose a method for generating a frequency-hopping sequence based on channel quality in a channel heterogeneous scenario to solve the technical problems of low channel rendezvous efficiency and low communication reliability existing in the prior art.

[0005] To achieve the above object, the technical solution of the present invention includes the following steps:

[0006] (1) Initialize the self-organizing network:

[0007] Initialize the self-organizing network in a channel heterogeneous scenario including N global channels with M channels being available channels, where N≥2, and the m-th available channel is H m ;

[0008] (2) Calculate the quality of each global channel and construct the pilot sequence:

[0009] Normalize the signal-to-interference-plus-noise ratio (SINR) of each global channel n and use the normalized SINR nr n of each global channel as the quality of this global channel to obtain a set including the qualities of N global channels; meanwhile, randomly select the index r of one available channel from M available channels, and obtain the coding string β according to r, and then construct a pilot sequence including the prefix string α, the random available channel index r, and the coding string β;

[0010] (3) Calculate the reference length P of the subsequence in the frequency hopping sequence to be generated;

[0011] (4) Construct a λ-type frequency hopping subsequence based on the channel quality:

[0012] Map the set of available channels based on the channel quality, and construct a λ-type frequency hopping subsequence with a length of L a including the indices of M available channels and different numbers of mapped channels;

[0013] (5) Construct an R-type frequency hopping subsequence through the pilot sequence;

[0014] (6) Obtain the generation result of the frequency hopping sequence:

[0015] Construct a frequency hopping matrix with four λ-type frequency hopping subsequences and R-type frequency hopping subsequences as columns respectively, and splice the rows in the frequency hopping matrix in sequence to obtain the frequency hopping sequence.

[0016] Compared with the prior art, the present invention has the following advantages:

[0017] The present invention generates a concatenated prefix string, a random available channel index, and a pilot sequence of a coding string by randomly selecting the channel index of an available channel, and splices the rows in a frequency hopping matrix constructed by four λ frequency hopping subsequences and an R-type frequency hopping subsequence with relatively prime lengths generated according to the pilot sequence in sequence to obtain a frequency hopping sequence, which can ensure that nodes successfully achieve channel convergence within a limited time, improve the convergence efficiency of nodes in the channel in an ad hoc network. At the same time, the signal-to-interference-plus-noise ratio (SINR) of each global channel after normalization is used as the quality of the global channel, and a λ-type frequency hopping subsequence is constructed by mapping the set of available channels based on the channel quality, fully considering the influence of heterogeneous interference on channel convergence, and effectively improving the convergence reliability of nodes in the channel in an ad hoc network. Description of the Drawings

[0018] Figure 1 is the implementation flowchart of the present invention.

[0019] Figure 2 is the comparison chart of the simulation results of the average convergence time, the maximum convergence time, and the convergence channel quality level between the present invention and the prior art.

[0020] Figure 3 is the comparison chart of the simulation results of the average convergence time, the maximum convergence time, and the convergence channel quality level between the present invention and the prior art. Specific Implementation Method

[0022] The present invention will be further described in detail below with reference to the drawings and specific embodiments:

[0023] Refer to Figure 1 , the present invention includes the following steps:

[0024] Step 1) Initialize the ad hoc network:

[0025] Initialize the ad hoc network in a heterogeneous channel scenario including N global channels and M available channels, where N≥2, and the m-th available channel is H m , in this embodiment, N = 100, M = 20;

[0026] All nodes in the ad hoc network in the channel heterogeneous scenario have different available channels.

[0027] Step 2) Calculate the quality of each global channel and construct a pilot sequence:

[0028] Normalize the signal-to-interference-plus-noise ratio (SINR) of each global channel n , and the normalized signal-to-interference-plus-noise ratio nr of each global channel nAs the quality of the global channel, a set including N global channel qualities is obtained; meanwhile, a pilot sequence including a prefix string α, a random channel index r, and a coding string β is constructed according to M available channels.

[0029] Normalized result nr n The calculation formula is:

[0030] nr n =(SINR n -SINR min ) / (SINR max -SINR min )

[0031]

[0032] where S n , I n , W n respectively represent the received signal power, interference signal power, and noise signal power of the nth channel, and SINR max , SINR min respectively represent the maximum and minimum values of all signal-to-interference-plus-noise ratios.

[0033] Normalize the signal-to-interference-plus-noise ratio of each global channel to the same range, and retain the advantage of the channels with larger signal-to-interference-plus-noise ratios in the selection process.

[0034] The implementation steps for constructing the pilot sequence are as follows:

[0035] (2a) Convert the index r of the available channel randomly selected from M available channels into a 6-bit binary code and encode it, then convert the encoded 8-bit binary code into a 4-bit quaternary code in reverse order to form the coding string β, where represents rounding up;

[0036] (2b) Use the 3-bit quaternary code 000 as the prefix string α, and cascade the prefix string α, the random available channel index r, and the coding string β in sequence to form the pilot sequence.

[0037] By adding a random channel index, it is ensured that the channels randomly selected by different nodes overlap within a certain period of time, thereby increasing the probability of rendezvous; encode the 6-bit binary code into an 8-bit binary code to maintain the DC balance of the communication system; convert the 8-bit binary code into a 4-bit quaternary code to ensure the mutual exclusivity and uniqueness of the length of the frequency-hopping subsequence, thereby ensuring the reliability of the algorithm; use the 3-bit quaternary code 000 as the prefix string α to solve the problem that the frequency-hopping sequences of two nodes in the clock asynchronous scenario cannot overlap due to cyclic shift.

[0038] Step 3) Calculate the reference length P of the subsequence in the frequency-hopping sequence to be generated:

[0039] The calculation formula for the reference length P of the subsequence in the frequency-hopping sequence to be generated is:

[0040] P = max(μ, 5)

[0041] where μ represents any prime number between M and 2M.

[0042] The reference length P of the subsequence in the frequency-hopping sequence to be generated is used to determine the basic repetition period of the frequency-hopping sequence, ensuring that the sequence has a certain length so that the node can cover all possible common channel combinations within a limited time.

[0043] Step 4) Construct a λ-type frequency-hopping subsequence based on the channel quality:

[0044] (4a) Perform x m times of mapping on the m-th available channel, and form a set including Q mapped channels with the mapping results of all available channels, where:

[0045]

[0046] where nr m represents the quality of the m-th available channel, and Σ represents the summation operation; the normalized channel quality is mapped to an extended channel set, so that the channels with higher channel quality have a larger number of channel mapping times.

[0047] (4b) Calculate the lengths L 0 , L 1 , L 2 and L 3 of the four λ-type frequency-hopping subsequences when the quaternary codes a = {0, 1, 2, 3} of the prefix string α and the coding string β in the pilot sequence are calculated according to the reference length P:

[0048]

[0049] (4c) Take M channels as the first M elements of each λ-type frequency-hopping subsequence, and take the channels with the most channel mapping times of P - M, P - M + 2, P - M + 4, and P - M + 3 as the other elements of the 1st, 2nd, 3rd, and 4th λ-type frequency-hopping subsequences respectively, to obtain four λ-type frequency-hopping subsequences.

[0050] In the heterogeneous channel and heterogeneous interference scenario, the available channels and interference levels of different nodes in the self-organizing network are different. Channel mapping reallocates channel priorities through the normalized channel quality, ensuring that the common channels between different nodes are preferentially selected as elements of the frequency hopping sequence, thereby improving the convergence efficiency; generating λ-type frequency hopping subsequences with relatively prime lengths based on the channel quality to increase the probability of nodes in the self-organizing network converging on the common channels with high channel quality.

[0051] Step 5) Construct an R-type frequency hopping subsequence through the pilot sequence:

[0052] Randomly reorder the M available channels, and select the indices of each channel after reordering a total of P times as the first P elements of the R-type frequency hopping subsequence. Then, use the random available channel index r as the (P + 1)-th and (P + 2)-th elements of the R-type frequency hopping subsequence, respectively, for a total of M times, to obtain an R-type frequency hopping subsequence with a length of L R = M×(P + 2).

[0053] The R-type subsequence can enumerate all common available channel combinations between different nodes by including M available channels and random available channel indices, so as to improve the channel convergence efficiency.

[0054] Step 6) Obtain the generation result of the frequency hopping sequence:

[0055] (6a) Calculate the number of rows K of the matrix according to the lengths of the four λ-type frequency hopping subsequences and the length of the R-type frequency hopping subsequence:

[0056] K = lcm(M(P + 2), P(P + 2)(P + 3)(P + 4))

[0057] where lcm(·) represents calculating the least common multiple;

[0058] (6b) Sort the λ-type frequency hopping subsequence with a = 0, that is, the first λ-type frequency hopping subsequence, the R-type frequency hopping subsequence, and the four λ-type frequency hopping subsequences by column. The first λ-type frequency hopping subsequence is sorted 3 times repeatedly, and each subsequence is repeatedly filled to K bits to obtain a frequency hopping matrix with a dimension of K×8. Then, splice the rows therein in sequence to obtain a frequency hopping sequence with a length of 8K.

[0059] Generate the frequency hopping subsequence in a matrix structure, decompose the complex frequency hopping sequence generation problem into a combination of multiple subsequences. The matrix structure allows generating various different frequency hopping sequences according to the prefix string, random available channel index, and coding sequence.

[0060] The following further illustrates the technical effects of the present invention in combination with simulation experiments:

[0061] 1. Simulation conditions:

[0062] Hardware platform for simulation experiments: The processor is an Intel i5-6400 CPU with a main frequency of 2.7 GHz and 16 GB of memory. Software platform for simulation experiments: Windows 10 operating system and Visual Studio 2019.

[0063] It is assumed that the two nodes in the self-organizing network are node a and node b respectively. C a represents the set of available channels of node a, and C b represents the set of available channels of node b. θ a = C a / N represents the proportion of the available channels of node a in the global channels. θ b = C b / N represents the proportion of the available channels of b in the global channels. C g = C a ∩C b represents the common available channels of the two nodes. θ g = C g / N represents the proportion of the common available channels in the global channels.

[0064] 2. Simulation content and result analysis:

[0065] Simulation 1: Comparative simulations are respectively carried out on the average convergence time, maximum convergence time, and convergence channel quality level between the present invention and the prior art. The results are as Figure 2 shown.

[0066] Simulation 2: Comparative simulations are respectively carried out on the average convergence time, maximum convergence time, and convergence channel quality level between the present invention and the prior art. The results are as Figure 3 shown.

[0067] Simulation 1: Set the global channel N = 100, θ a = 0.5, θ b = 0.5, θ g is increased from 0.05 to 0.5. Simulation experiments are respectively carried out between the present invention method and the prior art when the proportion of the common available channels in the global channels is between 0.05 and 0.5.

[0068] Simulation 2: Set θ a = 0.2, θ b = 0.3, θ g = 0.1. The global channel N is increased from 50 to 500. Simulation experiments are respectively carried out between the present invention method and the prior art when the common available channels are increased from 50 to 500.

[0069] The prior art refers to a frequency hopping sequence design method disclosed by the University of Electronic Science and Technology in "A Design Method of Anonymous Frequency Hopping Sequences for Multi-Antenna Cognitive Wireless Networks".

[0070] The average aggregation time ATTR, the maximum aggregation time MTTR, and the aggregation channel quality RI level are calculated as follows:

[0071]

[0072] ATTR = E[t(δ)]

[0073]

[0074] where t(δ) represents the aggregation time for any clock offset δ, E[·] represents taking the expectation, and nr f represents the quality of the f-th simulation aggregation channel, and F = 1000 represents the number of simulation experiments.

[0075] Referring to Figure 2 , where Figure 2 (a), Figure 2 (b), Figure 2 (c) are the average aggregation time, the maximum aggregation time, and the aggregation channel quality level respectively. From Figure 2 it can be seen that as the proportion of the common available channels in the global channels increases, the number of common channels for aggregation increases accordingly. The average aggregation time and the maximum aggregation time values of both methods show a downward trend. The average aggregation time and the maximum aggregation time of the present invention are lower than those of the prior art, showing an advantage in aggregation efficiency. Moreover, the aggregation channel quality RI level of the present invention is significantly higher than that of the prior art, indicating that the present invention can resist heterogeneous interference and ensure the reliability of the aggregation channel.

[0076] Referring to Figure 3 , where Figure 3 (a), Figure 3 (b), Figure 3 (c) are the average aggregation time, the maximum aggregation time, and the aggregation channel quality level respectively. From Figure 3 it can be seen that when the proportion of the common available channels and the proportion of nodes in the common available channels are the same, as the number of global channels increases, the number of common available channels increases synchronously, resulting in an upward trend in the average aggregation time and the maximum aggregation time of both methods. The average aggregation time and the maximum aggregation time of the present invention are lower than those of the prior art, and the aggregation channel quality level is significantly better than that of the prior art, indicating the advantages of the present invention in aggregation efficiency and the reliability of the aggregation channel.

Claims

1. A method for generating a frequency hopping sequence based on channel quality in a channel heterogeneous scenario, characterized in that: The steps include: (1) Initialize the self-organizing network: Initialize a self-organizing network in a channel heterogeneous scenario with N global channels and M channels as available channels, where N ≥ 2 and the mth available channel is H m ; (2) Calculate the quality of each global channel and construct a guiding sequence: The signal-to-interference-plus-noise ratio (SINR) of each global channel n Normalize and convert the normalized signal-to-noise ratio nr of each global channel into n As the quality of the global channel, a set of N global channel qualities is obtained; at the same time, an index r of an available channel among the M available channels is randomly selected, and a code string β is obtained according to r, and then a guide sequence including a prefix string α, a random available channel index r and a code string β is constructed; (3) Calculating the reference length P of the subsequence in the frequency hopping sequence to be generated; (4) Constructing a λ-type frequency hopping subsequence based on channel quality: Based on the channel quality, the available channel set is mapped, and a length L of indexes of M available channels and different numbers of mapped channels is constructed. a λ-type frequency hopping subsequence; (5) constructing an R-type frequency hopping subsequence through a pilot sequence; (6) Obtain the frequency hopping sequence generation result: A frequency hopping matrix is ​​constructed with four λ-type frequency hopping subsequences and an R-type frequency hopping subsequence as columns, and the rows in the frequency hopping matrix are sequentially concatenated to obtain a frequency hopping sequence.

2. The method according to claim 1, characterized in that The normalized result nr described in step (2) n , the calculation formula is: nr n =(SINR n -SINR min ) / (SINR max -SINR min ) Among them, S n ,I n , W n Respectively represent the received signal power, interference signal power, noise signal power of the nth channel, SINR max 、SINR min Respectively represent the maximum and minimum values ​​of all signal-to-interference-noise ratios.

3. The method according to claim 1, characterized in that The steps for constructing the guide sequence described in step (2) are as follows: (2a) Convert the index r of an available channel randomly selected from the M available channels into a 6-bit binary code After encoding, the 8-bit binary code is converted into a 4-bit quaternary code in order from back to front to form a code string β, where Indicates rounding up; (2b) The 3-bit quaternary code 000 is used as the prefix string α, and the prefix string α, the random available channel index r and the code string β are concatenated in sequence to form a pilot sequence.

4. The method according to claim 3, characterized in that The base length P of the subsequence in the frequency hopping sequence to be generated in step (3) is calculated as follows: P=max(μ,5) Wherein, μ represents any prime number between M and 2M.

5. The method according to claim 4, characterized in that The steps of constructing the λ-type frequency hopping subsequence based on the channel quality in step (4) are as follows: (4a) Perform x on the mth available channel m The mapping is repeated once, and the mapping results of all available channels are combined into a set of Q mapped channels, where: Among them, nr m represents the quality of the mth available channel, ∑ represents the summation operation; (4b) Calculate the lengths L0, L1, L2 and L3 of the four λ-type frequency hopping subsequences when the quaternary code a={0, 1, 2, 3} of the prefix string α and the code string β in the pilot sequence according to the reference length P: (4c) The M available channel indices are used as the first M elements of each λ-type frequency hopping subsequence, and the channel indices PM, P-M+2, P-M+4, and P-M+3 with the largest number of channel mapping times are used as the other elements of the 1st, 2nd, 3rd, and 4th λ-type frequency hopping subsequences, respectively, to obtain four λ-type frequency hopping subsequences.

6. The method according to claim 5, characterized in that The steps of constructing the R-type frequency hopping subsequence in step (5) are as follows: The M available channels are randomly reordered, and the index of each channel after reordering is selected P times as the first P elements of the R-type frequency hopping subsequence. Then, the random available channel index r is used as the P+1th and P+2th elements of the R-type frequency hopping subsequence respectively, and this is repeated M times to obtain a length of L. R =M×(P+2) R-type frequency hopping subsequence.

7. The method according to claim 6, characterized in that The step (6) of obtaining the frequency hopping sequence generation result, (6a) The number of rows K of the matrix is ​​calculated according to the lengths of the four λ-type frequency hopping subsequences and the length of the R-type frequency hopping subsequence: K=lcm(M(P+2),P(P+2)(P+3)(P+4)) Where lcm(·) means calculating the least common multiple; (6b) The λ-type frequency hopping subsequences with a=0, i.e., the first λ-type frequency hopping subsequence, the R-type frequency hopping subsequence, and the four λ-type frequency hopping subsequences are sorted by column, wherein the first λ-type frequency hopping subsequence is sorted 3 times, and each subsequence is repeatedly padded to K bits to obtain a frequency hopping matrix with a dimension of K×8, and then the rows thereof are concatenated in sequence to obtain a frequency hopping sequence with a length of 8K.

Images