A method for hiding information in music based on LoRa modulation

Abstract

The present invention discloses a method for broadcasting information to users using wireless information transmission technology without disturbing the users. LoRa modulation is used to embed information in music, so that users can use smart phones to obtain information while enjoying music. First, the present invention selects to use a frequency band near ultrasonic waves, uses LoRa to encode information and only plays the real part of the LoRa signal. Secondly, the present invention pre-processes the generated LoRa symbols respectively to ensure that the LoRa audio remains inaudible to the human ear. Then, frame synchronization is performed to determine the start of information transmission to ensure correct decoding of the smart phone. Finally, the decoding work with low bit error rate can be completed by comparing the starting frequency after point multiplication. The present invention has strong anti-interference ability, low bit error rate, wide transmission distance and coverage, and is easy to promote.

Description

Technical Field

[0001] The present invention relates to the technical field of wireless information transmission, and in particular to a method for hiding information in music based on LoRa coding to achieve sound wave transmission of a smart phone. Background Art

[0002] In daily life, it is very troublesome to broadcast information to users manually in public places such as shopping malls and restaurants without disturbing them. Using wireless information transmission technology to broadcast information can better solve this problem. Acoustic data transmission technology can encode information into music and realize wireless data transmission by playing music.

[0003] Existing acoustic data transmission technologies can be roughly divided into two categories. The first category is to encode data in a lower frequency band audible to the human ear. The main way to embed information in this type of coding is to modify the amplitude and frequency of the audio. This results in the encoded audio possibly changing the music quality of the original audio and thus affecting the user's auditory experience. The second category is to encode in a near-ultrasonic frequency band or ultrasonic frequency band that is inaudible to the human ear. The use of ultrasonic waves can well ensure the quality of music, but it requires a special ultrasonic speaker to play the audio. In addition, smartphones on the market cannot receive ultrasonic waves, making promotion difficult and costly.

[0004] Commonly used coding methods include frequency modulation and chirp coding. Frequency modulation is actually a modulation method that changes the carrier frequency according to the modulation signal. The frequency can be obtained by Fourier transform for decoding. Chirp is a coding method with linear frequency change. The traditional decoding method is to perform Hough transform or correlation decoding on the short-time Fourier transform of the sound. However, these two methods still produce noise that can be heard by the human ear during encoding even in the high-frequency band near ultrasonic waves, and both have a certain bit error rate. Therefore, it is not possible to achieve a low bit error rate transmission of information data to users without damaging the audio quality. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide a method for hiding information in music based on LoRa modulation, which can broadcast information data to users through smart phones without damaging the quality of music or disturbing users, and has low cost and is easy to promote.

[0006] In order to solve the above technical problems, the present invention provides a method for hiding information in music based on LoRa coding, comprising the following steps:

[0007] The present invention is divided into two parts: information sender and information receiver, wherein the information sender includes the following steps:

[0008] (1) The sender needs to prepare the music to be played and the information to be broadcast to the user, and encode the information to be sent into LoRa symbols; specifically,

[0009] (11) The sender uses 18KHz as the starting frequency, encodes the chirp in the frequency domain by cyclic translation, generates LoRa symbols, and only plays the real part of the LoRa signal;

[0010] (12) The sender encodes different data according to the different starting frequencies of the up-chirp, encoding the signal with a starting frequency of 18 kHz as 0, and encoding the signal with a starting frequency of 18.4 kHz after cyclic shift as 1;

[0011] (2) The sender needs to generate a start signal and add it before the information encoding LoRa symbol. The start signal is specifically a special fixed LoRa sequence;

[0012] (3) The sender must use a new window for each discontinuous frequency of the generated LoRa symbol to eliminate the noise caused by the frequency mutation. For the LoRa signal coded as 0 above, the following new window is used To perform preprocessing, the following new window is used for the LoRa signal coded as 1 above To perform preprocessing, where M represents the number of samples of a LoRa symbol, specifically expressed as M = fs*τ, where τ is the duration of a LoRa symbol, fs is the sampling rate, and n is an independent variable that increases from 0 to M. Each generated LoRa symbol is multiplied by w1(n) or w2(n) to obtain the preprocessed LoRa encoded audio, so that the frequency after the mutation at each frequency mutation point is gradually generated from 0, and the frequency before the mutation gradually disappears to 0;

[0013] (4) The sender mixes the generated processed LoRa symbol with the prepared music to be played to generate a mixed music file carrying the information, and uses a speaker to broadcast the music with hidden information;

[0014] The information receiver includes the following two steps:

[0015] (5) The receiver uses the smartphone’s microphone to receive the music with hidden information played by the speaker. First, the receiver needs to confirm the start of the frame.

[0016] (6) After the receiver confirms the start position of the frame, it will start decoding the LoRa signal; specifically:

[0017] (61) First, a down-chirp signal with a starting frequency of 800 Hz is generated, and each received LoRa symbol is processed by point multiplication of a down-chirp signal to obtain a single-frequency signal;

[0018] (62) Perform fast Fourier transform on the signal after point multiplication to obtain a single-frequency signal for distinguishing signals with different starting frequencies for decoding.

[0019] The steps (11) and (12) specifically include:

[0020] The encoding of bit 0 is generated by sampling from the following function at a given sampling rate: s(t) = cos(2πf0t + πμ / 2t 2 ), the specific encoding method of bit 1 is Where f0 is the starting frequency, f c is the intermediate frequency, μ is defined as the frequency change rate, and τ is the duration of a LoRa symbol.

[0021] The step (5) specifically comprises the following steps:

[0022] (51) The receiver uses short-time Fourier transform detection to filter out the high-frequency bands of the original music and environmental noise to align the symbol data;

[0023] (52) When the receiver detects that the audio frequency is close to the starting frequency of LoRa, it will enter a special state. In this state, the present invention attempts to decode the upcoming fixed sequence of LoRa start signals. When all the following fixed sequence of LoRa symbols are successfully decoded, frame synchronization is achieved. Otherwise, the window is slid by one sample and the next attempt is started.

[0024] (53) After the frame synchronization is completed and the time point when the signal starts is successfully located, the present invention will move backward ten units each time to attempt 50 decoding attempts and record the decoding results, and finally compare and output the decoding result with the lowest bit error rate.

[0025] The step (62) specifically includes:

[0026] The receiver first generates a down-chirp signal with a starting frequency of 800 Hz. The specific generation method is: s(t) = cos(2πf1t-πμ / 2t 2 ), where f1 is 800Hz, μ is defined as the frequency change rate, and then the received LoRa symbol is multiplied with this down-chirp to obtain a new signal. The energy and energy of the new signal in different frequency bands are calculated to determine the specific calculation method of the starting frequency: Let mag(i) be the amplitude of frequency i, then If e1>e2, the symbol is decoded as bit 0, otherwise it is decoded as bit 1.

[0027] Beneficial effects: Compared with the prior art, the present invention has the following advantages:

[0028] (1) The audio embedded with information will not reduce the quality of the music, and users will not be aware of the existence of the LoRa symbol while listening to music. In other words, users can receive LoRa information using their smartphones while enjoying music.

[0029] (2) Directly broadcast the information to be transmitted using audio, which saves labor while sending information conveniently and quickly without disturbing the user.

[0030] (3) Compared with traditional frequency modulation and chirp coding, the present invention has strong anti-interference ability and strong resistance to multipath and Doppler effects, and has a lower bit error rate, a longer transmission distance and a wider coverage range. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 Flowchart of the present invention

[0032] Figure 2 The experimental diagram of the bit error rate of the present invention changing with the coding method

[0033] Figure 3 This is the experimental diagram of the bit error rate changing with volume of the present invention

[0034] Figure 4 This is the experimental diagram of the bit error rate of the present invention changing with environmental noise DETAILED DESCRIPTION

[0035] The present invention is specifically introduced below.

[0036] When the present invention is implemented, the information sender needs a speaker, which is used to broadcast the audio file mixed with the LoRa encoded information to send the information. The information receiver needs a smart phone, which is used to receive the mixed audio file and then decode it to receive the information.

[0037] The whole process is divided into two parts: the information sender and the information receiver. The information sender includes the following steps:

[0038] (1) The sender needs to prepare the music to be played and the information to be broadcast to the user, and encode the information to be sent into LoRa symbols.

[0039] The present invention uses LoRa to encode information, and uses chirp symbols in the same frequency band but with different starting frequencies to represent 0 / 1 bits. Let f0 be the minimum frequency of the LoRa symbol in the frequency band, and f1 be the maximum frequency of the LoRa symbol in the frequency band, is the middle frequency between f0 and f1. τ is the duration of a LoRa symbol. The present invention recommends setting τ to 100ms. μ is the frequency change rate, specifically

[0040] Then, the LoRa symbol is obtained by time sampling two cosine waves with a duration of τ. The present invention adopts the encoding method of LoRa, assuming f0 is the starting frequency of bit 0, is the starting frequency of 1 bit, and the present invention sets f0 to 18KHz. The present invention sets the 0-bit cosine wave to s(t)=cos(2πf0t+πμ / 2t 2 ), set the cosine wave of bit 1 to That is, the frequency of bit 0 increases directly from f0 to f1, while the frequency of bit 1 increases first from f c Increase to f1, and then increase from f0 to f c .

[0041] (2) The sender needs to generate a start signal and add it before the information encoding LoRa symbol. The start signal is specifically a special fixed LoRa sequence.

[0042] Since the client cannot know the exact time when the encoding starts, direct decoding will cause a large bit error rate. Before the LoRa information encoding described in (1), the present invention needs to add a start signal to confirm the exact start time of the frame for correct decoding. To this end, the present invention generates a special LoRa sequence of fixed length at the beginning of the information encoding as a start signal.

[0043] (3) The sender must use a new type of windowing processing for each discontinuous frequency of the generated LoRa symbol to eliminate the noise caused by frequency mutations.

[0044] The present invention uses the following new window for the LoRa signal with the above code 0 To perform preprocessing, the following new window is used for the LoRa signal coded as 1 above To perform preprocessing, where M represents the number of samples of a LoRa symbol, specifically expressed as M = fs*τ. Where fs is the sampling rate, τ is the duration of a chirp symbol, and n is an independent variable that increases from 0 to M.

[0045] The present invention multiplies each generated LoRa symbol by the corresponding w(n) to obtain the preprocessed LoRa audio, so that each LoRa symbol including the discontinuous frequency inside each LoRa symbol is gradually generated from 0 and gradually disappears before the frequency suddenly changes. At this time, each LoRa symbol can be seamlessly connected to the next LoRa symbol and the LoRa symbol corresponding to 1 bit can also be connected without frequency sudden change.

[0046] (4) The sender mixes the generated processed LoRa symbol with the prepared music to be played to generate a mixed music file carrying the information, and uses a speaker to broadcast the music with hidden information.

[0047] The information receiver includes the following steps:

[0048] (5) The receiver uses the smartphone’s microphone to receive the music with hidden information played by the speaker, and first confirms the start of the frame.

[0049] (51) The receiver uses short-time Fourier transform to detect frequencies above 17500 Hz in the music to filter out the original music and environmental noise to align the symbol data.

[0050] (52) When the receiver detects that the audio frequency is close to the starting frequency of LoRa, it will enter a special state. In this state, the present invention attempts to decode the upcoming fixed sequence of LoRa start signals. When all the following fixed sequence of LoRa symbols are successfully decoded, frame synchronization is achieved. Otherwise, the window is slid by one sample and the next attempt is started.

[0051] (6) After the receiver confirms the start position of the frame, it will start decoding the LoRa signal.

[0052] (61) Assume that the sound corresponding to the LoRa symbol is s0. The present invention performs a fast Fourier transform on s0 to obtain the amplitude of the frequency, and sets mag(i) as the amplitude of frequency i.

[0053] (62) Calculate the energy and

[0054] (63) Compare the size of e1 and e2. If e1>e2, decode the symbol as 0 bit, otherwise decode it as 1 bit. Record the bit error rate at this time and shift the entire audio backward by 10 samples.

[0055] (64) After moving 10 samples, a new s0 is obtained. The above decoding steps are repeated 50 times. A set of decoded data with the lowest bit error rate is obtained by comparison and output. At this point, the receiver completes the decoding work and obtains the information hidden in the music.

[0056] The effect of the present invention can be further illustrated by the following experiments.

[0057] Experiment 1 tested the bit error rate comparison between the direct frequency modulation coding method, the traditional chirp coding and the LoRa coding proposed in this invention under the same environment. We set the volume to 100% and placed the smartphone at the same distance from the speaker to test the bit error rate. The experimental results are shown in the attached figure. Figure 2 The experimental results show that the present invention performs best among several coding algorithms and has the lowest bit error rate.

[0058] Experiment 2 tested the effect of different speaker volumes on the bit error rate. The volume was set to 10%, 30%, 50%, 70% and 100% of the speaker broadcast volume to measure the bit error rate. The experimental results are shown in the attached figure. Figure 3 The experimental results show that the present invention is quite robust, and the bit error rate increases only when the volume is below a lower threshold. In most other cases, the present invention is stable.

[0059] Experiment 3 tested the effect of different decibels of environmental noise on decoding. The experiment controlled the environmental noise to 30 decibels, 40 decibels, 50 decibels and 60 decibels respectively, and the transmission data decoding was carried out under these conditions. The experimental results are shown in the attached Figure 4 As shown. It can be concluded from the experiment that the present invention has strong anti-interference ability and can operate well under extremely bad environmental noise conditions. This means that the present invention can be applied to various types of places, whether it is a quiet coffee shop or a noisy shopping mall, the present invention can work well.

Claims

1. A method for hiding information in music based on LoRa modulation, which is divided into two parts: information sender and information receiver, and is characterized by: The sender of the message includes the following four steps: (1) The sender needs to prepare the music to be played and the information to be broadcast to the user, and encode the information to be sent into LoRa symbols; specifically, (11) The sender uses 18KHz as the starting frequency, encodes the chirp in the frequency domain by cyclic translation, generates LoRa symbols, and only plays the real part of the LoRa signal; (12) The sender encodes different data according to the different starting frequencies of the up-chirp, encoding the signal with a starting frequency of 18 kHz as 0, and encoding the signal with a starting frequency of 18.4 kHz after cyclic shift as 1; (2) The sender needs to generate a start signal and add it before the information encoding LoRa symbol. The start signal is specifically a special fixed LoRa sequence; (3) The sender must use a new window for each discontinuous frequency of the generated LoRa symbol to eliminate the noise caused by the frequency mutation. For the LoRa signal coded as 0 above, the following new window is used To perform preprocessing, the following new window is used for the LoRa signal coded as 1 above To perform preprocessing, where M represents the number of samples of a LoRa symbol, specifically expressed as M = fs*τ, where τ is the duration of a LoRa symbol, fs is the sampling rate, and n is an independent variable that increases from 0 to M. Each generated LoRa symbol is multiplied by w1(n) or w2(n) to obtain the preprocessed LoRa encoded audio, so that the frequency after the mutation at each frequency mutation point is gradually generated from 0, and the frequency before the mutation gradually disappears to 0; (4) The sender mixes the generated processed LoRa symbol with the prepared music to be played to generate a mixed music file carrying the information, and uses a speaker to broadcast the music with hidden information; The information receiver includes the following two steps: (5) The receiver uses the smartphone’s microphone to receive the music with hidden information played by the speaker. First, the receiver needs to confirm the start of the frame. (6) After the receiver confirms the start position of the frame, it will start decoding the LoRa signal; specifically: (61) First, a down-chirp signal with a starting frequency of 800 Hz is generated, and each received LoRa symbol is processed by point multiplication of a down-chirp signal to obtain a single-frequency signal; (62) Perform fast Fourier transform on the signal after point multiplication to obtain a single-frequency signal for distinguishing signals with different starting frequencies for decoding.

2. The method for hiding information in music based on LoRa modulation according to claim 1, characterized in that: In steps (11) and (12) of the information sender, the encoding of bit 0 is generated by sampling from the following function at a given sampling rate: s(t)=cos(2πf0t+πμ / 2t 2 ), the specific encoding method of bit 1 is Where f0 is the starting frequency, f c is the intermediate frequency, μ is defined as the frequency change rate, and τ is the duration of a LoRa symbol.

3. The method for hiding information in music based on LoRa modulation according to claim 1, characterized in that: In step (5) of the information receiver, the receiver uses the microphone of the smart phone to receive the music played by the speaker. In order to correctly decode, the beginning of the frame must be confirmed first, which specifically includes the following steps: (51) The receiver uses short-time Fourier transform detection to filter out the high-frequency bands of the original music and environmental noise to align the symbol data; (52) When the receiver detects that the audio frequency is close to the starting frequency of LoRa, it will enter a special state. In this state, it will try to decode the upcoming fixed sequence of LoRa start signals. When all the following fixed sequence of LoRa symbols are successfully decoded, frame synchronization is achieved. Otherwise, the window will slide one sample and start the next attempt. (53) After the frame synchronization is completed and the time point when the signal starts is successfully located, the decoding will be attempted 50 times by moving backward ten units each time and recording the decoding results. Finally, the decoding result with the lowest bit error rate will be compared.

4. The method for hiding information in music based on LoRa modulation according to claim 2, characterized in that: In step (62) of the information receiver, the receiver first generates a down-chirp signal with a starting frequency of 800 Hz, and the specific generation method is: s(t) = cos(2πf1t-πμ / 2t 2 ), where f1 is 800Hz, μ is defined as the frequency change rate, and then the received LoRa symbol is multiplied with this down-chirp to obtain a new signal. The energy and energy of the new signal in different frequency bands are calculated to determine the starting frequency. The specific calculation method is: let mag(i) be the amplitude of frequency i, let If e1>e2, the symbol is decoded as bit 0, otherwise it is decoded as bit 1.

Images