A spectrum sensing and trusted sharing method based on blockchain

Abstract

The present invention discloses a spectrum sensing and trusted sharing method based on blockchain, which is applied to the field of blockchain and cognitive radio. In view of the problem that dishonest users interfere with the normal spectrum sensing and sharing process for their own interests in the spectrum sharing mechanism based on cognitive radio in the existing wireless network, the present invention designs a spectrum sensing and trusted sharing method based on blockchain. In an untrusted environment, the accurate spectrum sensing mechanism can be used to realize the user's accurate judgment on the busy and idle status of the channel, and the consensus mechanism based on the perception accuracy can be used to ensure the user's credibility and efficiency in using the idle channel, thereby realizing the accuracy, credibility and efficiency of spectrum sensing and sharing.

Description

Technical Field

[0001] The present invention belongs to the field of blockchain and cognitive radio, and in particular relates to an authentication technology for spectrum sensing and spectrum sharing. Background Art

[0002] The rapid development of wireless communication in the field of research and application has brought about a surge in the number of users and the access of massive devices. The demand for interconnection has caused an exponential increase in the demand for radio spectrum resources, while the carrier of wireless communication - radio spectrum resources are very expensive and limited. In order to effectively solve the problem of shortage of spectrum resources, cognitive radio technology has been proposed and studied and applied. Cognitive radio technology is to sense the spectrum signals of authorized users through unauthorized users. When the channel is perceived to be idle, the unauthorized user can borrow the idle spectrum resources for communication without hindering the normal communication of the authorized user. Using cognitive radio technology, spatial reuse of spectrum resources can be achieved when the authorized spectrum is idle, effectively alleviating the problem of spectrum shortage and improving the utilization rate of spectrum resources.

[0003] However, there are still some problems in the application of cognitive radio technology. As the scale of the network becomes larger and larger, unauthorized users accessing the wireless network are mostly from different parties, and the purpose of accessing the wireless network is also different. They have independent decision-making methods and behavioral capabilities in the network, and there is a great deal of competition when using the wireless network. Therefore, when applying cognitive radio technology that senses the spectrum resources of authorized users and shares them when idle, there are some unauthorized users who want to reduce the efficiency of spectrum resource utilization, disrupt the normal operation of the network, or deceive other unauthorized users to exclusively enjoy idle channels. These dishonest users deliberately publish false perception information during spectrum sensing to mislead the judgment of other unauthorized users in the network on the busy and idle status of the channel, and use unfair means to compete for the right to use idle channels during spectrum sharing, which greatly reduces the fairness and efficiency of using cognitive radio technology to solve the problem of spectrum resource shortage. Therefore, the effective operation of cognitive radio technology needs to be based on accurate spectrum sensing and reliable spectrum sharing.

[0004] Currently, there are two ways to manage spectrum sensing and spectrum sharing: authentication and non-authentication. The authentication method usually relies on a centralized management environment, and transmits and stores spectrum sensing information and spectrum sharing results through a central node. Using the authentication method in a wireless network will bring additional overhead and single point failure problems. Therefore, in a distributed scenario where unauthorized users in a wireless network belong to different users, it is not suitable to use the authentication method to manage spectrum sensing and spectrum sharing. The non-authentication method can build a distributed and trusted spectrum sensing and sharing environment in a wireless network by leveraging the advantages of blockchain technology, such as decentralization, immutability, traceability, anonymity, and transparency. However, the non-authentication method also has certain problems. On the one hand, in order to achieve accurate spectrum sensing, all unauthorized users in the wireless network participate in collaborative spectrum sensing. Then, the accuracy of the collaborative sensing results in judging whether the authorized user channel is idle will be the highest, and the utilization rate of idle spectrum in the network is likely to reach the maximum. However, compared with not participating in spectrum sensing, participating in spectrum sensing requires additional energy consumption, and all unauthorized users in the network participating in spectrum sensing will generate the highest additional energy consumption. Therefore, it is necessary to improve the accuracy of spectrum sensing while limiting the additional energy consumption in the spectrum sensing process, thereby improving the overall performance of the wireless network. On the other hand, in order to achieve efficient spectrum sharing, the blockchain consensus mechanism can be used to maintain the security and consistency of spectrum sharing in a distributed system that does not require mutual trust. However, in an untrusted spectrum sharing environment, in order to reach a consensus on unauthorized users who can use idle spectrum, the most widely used POW consensus algorithm currently only relies on the computing power of unauthorized users and cannot effectively reduce the impact of dishonest users. Therefore, to achieve efficient spectrum sharing, a consensus algorithm that can effectively distinguish between honest and dishonest users is needed to ensure fairness and efficiency in the use of idle authorized spectrum resources. Summary of the invention

[0005] In order to solve the above technical problems, the present invention proposes a spectrum sensing and trusted sharing method based on blockchain, which can efficiently and reliably realize accurate spectrum sensing and efficient spectrum sharing in an untrusted environment, which is conducive to solving the problem of increasingly scarce wireless spectrum resources and promoting the development of cognitive radio, and has important research significance and application value.

[0006] The technical solution adopted by the present invention is: a spectrum sensing and trusted sharing method based on blockchain, based on the application scenario including: primary user PU, secondary user SU, in an untrusted network environment, each secondary user SU competes to use the idle channel of the primary user PU; the specific sharing method includes the following steps:

[0007] S1. Each secondary user SU broadcasts whether it has data transmission requirements through the control channel. If the i-th secondary user SU participates in spectrum sensing, it also broadcasts its own spectrum sensing value SNRi (t);

[0008] S2. All secondary users obtain the collaborative spectrum sensing results of the entire network based on their own spectrum sensing values ​​and the spectrum sensing values ​​received from other secondary users.

[0009] S3. Each secondary user updates its own integral value and reputation value according to the absolute value of the difference between its own spectrum sensing value and the collaborative spectrum sensing result of the whole network; and broadcasts its own updated integral value and reputation value through the control channel;

[0010] S4. If there is a secondary user SU whose credit value is less than the basic credit value, the secondary user SU loses the qualification to apply for an idle channel in this round;

[0011] S5. If the cooperative spectrum sensing result of the entire network in this time slot is greater than the threshold value for judging whether the PU channel is busy or idle, it is considered that the current channel of the primary user PU is not idle, and step S7 is executed; otherwise, step S6 is executed;

[0012] S6, the secondary user SU with the highest integral value in this time slot obtains the right to generate a block in this time slot and use the idle channel of the primary user PU. The secondary user SU uses the idle channel of the primary user PU in this time slot to send the data to be sent. After the data is successfully sent, the integral value of this time slot is deducted from the sending cost to obtain its integral value of this time slot; return to step S1 to compete for the idle channel of the primary user PU in the next time slot;

[0013] S7. The secondary user SU with the highest integral value in this time slot will form a block for this time slot based on the collaborative spectrum sensing result of the entire network in this time slot, the spectrum sensing value of each secondary user SU, and the latest integral value and reputation value of each secondary user SU in this time slot, and write it into the blockchain; return to step S1 to compete for the idle channel of the primary user PU in the next time slot.

[0014] Beneficial effects of the invention: The method of the invention has the following advantages:

[0015] 1. The blockchain-based spectrum sensing and sharing network architecture proposed in the present invention can effectively improve the utilization rate of idle spectrum in an untrusted wireless network environment. By reasonably modeling the behaviors of PU, HSU, and DSU and considering the influence and constraints of mutual decision-making, a blockchain-based spectrum sensing and spectrum sharing solution can be realized in an untrusted environment.

[0016] 2. The accurate spectrum recognition mechanism in an untrusted environment proposed in the present invention can effectively improve the accuracy of spectrum recognition in a distributed, untrusted network environment; through the collaborative spectrum sensing of multiple SUs, dynamically analyzing the probability of SUs participating in collaborative spectrum sensing, avoiding DSUs from interfering with the results of collaborative spectrum sensing, and comprehensively considering the additional energy consumption of participating in spectrum sensing, the compromise between spectrum recognition accuracy and energy consumption is achieved with perception accuracy as the guide;

[0017] 3. The lightweight consensus mechanism based on perception accuracy proposed in the present invention can effectively ensure the efficiency and credibility of users in using idle channels; by proposing a lightweight consensus algorithm based on spectrum perception accuracy and using integral value and reputation value constraints, the impact of DSU is reduced in an untrustworthy spectrum sharing environment, and idle spectrum sharing is achieved fairly and efficiently. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a spectrum sensing and sharing network architecture based on blockchain;

[0019] Figure 2 It is a time slot division diagram.

[0020] Figure 3 This is a schematic diagram of the scoring trend.

[0021] Figure 4 It is a schematic diagram of the cumulative trend of points value / credit value. DETAILED DESCRIPTION

[0022] To facilitate those skilled in the art to understand the technical content of the present invention, the present invention is further explained below with reference to the accompanying drawings.

[0023] Since the capabilities, positions and needs of SUs in wireless networks to participate in spectrum sensing and sharing are different, an untrustworthy spectrum sensing and sharing environment is formed. The "untrustworthy environment" mentioned here mainly refers to behaviors such as uneven spectrum sensing capabilities of various nodes and nodes that may covertly publish misleading spectrum sensing results due to their own communication needs, but does not include non-covert malicious attack behaviors, such as channel interference, forced occupation, etc. The effectiveness and efficiency of spectrum sharing technology based on cognitive radio focus on the accuracy of spectrum sensing, and the more SUs participate in spectrum sensing to improve the accuracy of spectrum sensing, the more additional energy consumption problems are faced, and at the same time, when DSU exists, the consensus mechanism of blockchain is used to fairly, safely, efficiently and reasonably use idle spectrum resources. The present invention proposes a spectrum sensing and trusted sharing method based on blockchain, which includes an accurate spectrum sensing mechanism in an untrustworthy environment and a lightweight consensus mechanism based on sensing accuracy. In an untrustworthy environment, the spectrum sensing mechanism is used to realize the user's accurate judgment of the busy and idle state of the channel, and the consensus mechanism is used to ensure the credibility and efficiency of the user's use of idle channels, thereby realizing fair, reasonable and efficient spectrum sharing. The following is a specific description of the technical solution:

[0024] 1. Spectrum sensing and shared network architecture based on blockchain

[0025] like Figure 1 As shown in the figure, a spectrum sensing and sharing network architecture based on blockchain is proposed to solve the needs of accurate spectrum sensing and efficient spectrum sharing in an untrusted network environment in a distributed, secure and reliable manner, effectively improve the utilization rate of idle spectrum, and minimize the impact of DSU on network performance in the process.

[0026] In the network, PU (Primary User) is the only legally authorized user, and SU (Second User, SU) is an unauthorized user. Their transmissions are all time slotted. The time slot division method is as follows: Figure 2 The behavior of each user is described as follows:

[0027] (1)PU

[0028] In this architecture, PU is a formal authorized user with a license and can choose to use any time slot in the channel. At the same time, PU's use of a time slot must be at the beginning of the time slot, and PU cannot start using the channel in the middle of a time slot.

[0029] The energy signal of the PU in the busy state can be sensed by any SU in the network. When the PU has occupied the channel at the beginning of a time slot, no SU can use the channel. If an SU occupies the channel at this time, it will seriously interfere with the normal communication of the PU. This illegal occupation can be divided into two cases: (1) The SU is a spectrum attack node and deliberately interferes with the normal communication of the PU. This situation belongs to the scope of network security management and anti-interference communication, which exceeds the scope of "untrusted environment" defined in the present invention and is not within the scope of consideration of the present invention. (2) The SU is not an attack node, but it may mistakenly occupy the channel being used by the PU due to inaccurate channel perception. In this case, the specific mechanism of the present invention - the accurate spectrum perception mechanism in an untrusted environment and the lightweight consensus mechanism based on perception accuracy can be used to achieve accurate perception and fair allocation of spectrum status, and avoid channel misoccupancy and disorderly occupation.

[0030] (2)SU

[0031] SU communicates by borrowing the idle spectrum resources of PU. Since there is no direct communication information exchange between SU ​​and PU, each SU must and only sense whether the PU channel is idle during the spectrum sensing phase in a time slot. If the spectrum sensing fusion result is that the PU channel is idle, the SU will select the SU that can use the PU idle spectrum for communication in this time slot according to the lightweight consensus mechanism based on perception accuracy proposed by the present invention; if the spectrum sensing fusion result is that the PU channel is busy, it will wait until the spectrum sensing phase of the next time slot to sense the busy and idle state of the PU again.

[0032] SUs can affect each other and can also communicate with each other through control channels. When the PU channel is idle, only one SU can obtain the right to use the spectrum in a time slot after competition among SUs to avoid mutual interference between SUs. SUs can broadcast their own spectrum perception values, whether data transmission is required in this time slot, and other information through the control channel. The control channel is reliable, and the information transmitted in it will not be lost or tampered with. At the same time, the time for transmitting information using the control channel can be ignored.

[0033] Considering that SUs come from different parties, there is a clear competition relationship in the use of spectrum. In a time slot, at most only one SU can get the opportunity to transmit data packets when the PU is idle. Some SUs may even have dishonest behaviors to reduce the utilization efficiency of spectrum resources, disrupt the normal operation of the network, or deceive other SUs to monopolize idle channels. Based on this, the present invention divides SUs into HSU (Honest SU) and DSU (Dishonest SU).

[0034] 1) HSU

[0035] HSU strictly adheres to the regulations of the wireless network, performs spectrum sensing according to the system operation mode, judges the busy and idle states of the PU channel, and decides the SU that can utilize the idle spectrum based on the consensus mechanism when the channel is idle. Among HSU, there are only differences in spectrum sensing capabilities caused by factors such as the distance from the PU and the surrounding channel environment. The sensing accuracies of different HSU in the process of spectrum sensing are high or low, but they will not deliberately perform dishonest behaviors and publish false spectrum sensing information.

[0036] At the beginning of a time slot, HSU will decide whether to perform spectrum sensing according to the probability P a and then fuse the spectrum sensing results with other SUs. When the PU is idle, based on the lightweight consensus mechanism based on sensing accuracy proposed in the present invention, the SU that can utilize the idle spectrum is decided. This SU records all the spectrum sensing results of the SUs and the corresponding spectrum usage situation in this time slot into a new block, writes the new block into the blockchain, and simultaneously accesses the idle channel of the PU to conduct data packet transmission communication.

[0037] 2) DSU

[0038] In order to achieve the purpose of reducing the utilization efficiency of spectrum resources, disrupting the normal operation of the network, or deceiving other SUs to exclusively occupy the idle channel, DSU will actively perform spectrum sensing at the beginning of each time slot, and flexibly choose to publish the modified self-sensing result according to different sensing results to affect the fusion of sensing results, or not publish or honestly publish its own sensing result to hide its identity as DSU. Based on the foregoing description of the "untrusted environment", DSU will not exhibit non-concealed malicious attack behaviors such as channel interference and forced occupation. That is, DSU will neither deliberately emit energy signals to directly interfere with the spectrum sensing result when the rest of the HSU perform spectrum sensing, nor transmit its own data packets or send interference signals when an SU conducts data packet transmission during the spectrum usage stage. The dishonest behavior of DSU may be completed by unconscious cooperation among DSU, or when necessary, DSU negotiate dishonest behaviors to prevent the dishonest behaviors of different DSU from offsetting each other. Assume that the spectrum sensing value of the i-th DSU in the t-th time slot is SNR i (t), and its threshold for judging the busy and idle states of the PU channel is ε i (t). When |SNR i (t) - ε i (t)| < d, DSU will modify its true spectrum sensing value to interfere with the fusion process of the sensing results among SUs: if SNR i (t) - ε i (t) < d, DSU will reduce its own spectrum sensing value to to satisfy Equation (1) and broadcast it to other SUs;

[0039]

[0040] where n is the number of SUs in the network, and n2 is the number of DSU in the network, is the probability that each HSU participates in spectrum sensing estimated by the DSU, and its calculation formula is:

[0041]

[0042] where N s is the total number of SUs participating in spectrum sensing per time slot obtained by the DSU, which can be obtained by averaging the total number of SUs participating in spectrum sensing in any several time slots selected by the DSU.

[0043] If ε i (t) - SNR i (t) < d, the DSU increases its own spectrum sensing value to to satisfy Equation (3) and broadcast it to other SUs.

[0044]

[0045] The purpose of doing this is to deceive the HSU to increase the opportunity for DSU to exclusively use the idle channel to transmit messages, or to interfere with the HSU's use of idle spectrum resources, reduce the network performance, and even disrupt the normal operation of the network.

[0046] The parameter d takes corresponding appropriate empirical values according to the specific network environment and application scenario. For example, in the scenario of sharing the 4G idle frequency band, the typical value of d is 1dB. When |SNR i (t) - ε i (t)| ≥ d, for the DSU to tamper with the spectrum sensing fusion result, it needs to modify its spectrum sensing result to a value that is extremely deviated from the normal sensing result, which will expose the identity of the DSU. At this time, the DSU has two behavioral modes:

[0047] a) Do not participate in the fusion of sensing results in this time slot, that is, do not broadcast spectrum sensing information to disguise itself as if the HSU will perform spectrum sensing with a certain probability;

[0048] b) Honestly publish spectrum sensing information to reduce the risk of being judged as a DSU in the consensus process. In a certain time slot, the choice of the two behavioral modes of the DSU will be determined by the behavior of the DSU in the previous time slot. If the DSU completed dishonest behavior in the previous time slot, then it chooses behavior b in this time slot to offset the negative evaluation of its dishonest behavior; if the DSU completed behavior a or behavior b in the previous time slot, then it chooses behavior a in this time slot.

[0049] 2. Accurate spectrum sensing mechanism in untrusted environments

[0050] At the beginning of each time slot, all HSUs start with P a The probability of HSU choosing to participate in spectrum sensing in this time slot is a i (t) = 1 means that HSU chooses not to participate in spectrum sensing in this time slot. i (t) = 0, then whether the HSU participates in spectrum sensing can be expressed by formula (4):

[0051]

[0052] The probability that the PU spectrum is idle in each time slot is P0, and each SU uses the theory of energy sensing to perform spectrum sensing. i (t) represents the signal-to-noise ratio of the PU signal perceived by the i-th SU in the t-th time slot, which can be expressed by formula (5):

[0053]

[0054] in, They represent the channel fading power, PU signal power and channel noise power sensed by the i-th SU in the t-th time slot respectively. Then the probability of PU spectrum idleness and false alarm probability obtained by SU based on energy perception can be expressed by equations (6) and (7):

[0055]

[0056]

[0057] in, N(t) is the number of SUs participating in spectrum sensing at the tth time slot, ε i (t) is the threshold value of energy perception of the i-th SU in the t-th time slot, which can be expressed by formula (8):

[0058]

[0059] The weighted arithmetic average method is used for the fusion of perception results. The weight is used to measure the credibility of a SU in the entire system network to reduce the impact of SUs that show dishonesty in the previous time slot on the fusion of perception results. The weight is maintained according to the credibility of each SU. The fusion result in the tth time slot shows the probability of PU spectrum idleness, the probability of false alarm, the signal-to-noise ratio of the PU signal after fusion, and the threshold value of energy perception can be expressed by formula (9), formula (10), formula (11) and formula (12):

[0060]

[0061]

[0062]

[0063]

[0064] where, ω i (t) is the weight of the i-th SU in the t-th time slot during sensing result fusion, which can be expressed by Equation (13):

[0065]

[0066] where, is the reputation value of the i-th SU in the t-th time slot. When t = 0, When t ≠ 0, is the reputation value obtained by the i-th SU in the t-th time slot. The value-taking method is as follows: In the t-th time slot of the lightweight consensus mechanism based on sensing accuracy, if the signal-to-noise ratio of the PU signal after sensing result fusion is greater than the threshold of energy sensing, it is considered that the PU channel is in a busy state; if the signal-to-noise ratio of the PU signal after sensing result fusion is less than the threshold of energy sensing, it is considered that the PU channel is in an idle state. When the sensing result fusion determines that the PU is idle, it is represented by e(t) = 1, and when the sensing result fusion determines that the PU is busy, it is represented by e(t) = 0. Then, whether the PU is in an idle state can be represented by Equation (14):

[0067]

[0068] In a time slot, when the SU performs spectrum sensing result fusion, the DSU will make a decision on whether to tamper with the sensing result in this time slot according to its own energy sensing value of the PU. When SNR i (t) - ε i (t) < d, the DSU reduces its own spectrum sensing value to to try to mislead the spectrum fusion result from busy to idle; when ε i (t) - SNR i (t) < d, the DSU increases its own spectrum sensing value to to try to mislead the spectrum fusion result from idle to busy. When the sensing result fusion of all HSU participating in spectrum sensing in the t-th time slot determines that the PU channel is idle, it is represented by H(t) = 1, and when it determines that the PU channel is busy, it is represented by H(t) = 0. When the sensing result fusion with the participation of the DSU in the t-th time slot determines that the PU channel is idle, it is represented by D(t) = 1, and when it determines that the PU channel is busy, it is represented by D(t) = 0. Then, there are four cases of the DSU performing dishonest behavior, which can be represented by Equation (15):

[0069] H(t)=1,D(t)=1 (15-a)

[0070] H(t)=1,D(t)=0 (15-b)

[0071] H(t)=0,D(t)=1 (15-c)

[0072] H(t)=0,D(t)=0 (15-d)

[0073] The probability P of DSU attempting to tamper with the spectrum sensing result fusion, but the tampering fails r It can be expressed by formula (16):

[0074]

[0075] The perception result fusion is not affected by DSU and remains normal, which is represented by r(t)=1. The perception result fusion is affected by DSU and changes, which is represented by r(t)=0. Then, whether the spectrum sensing result fusion is affected by DSU can be expressed by formula (17):

[0076]

[0077] In the tth time slot, if the following conditions are met at the same time: ① the PU channel is idle; ② no false alarm occurs in the fusion of the sensing results; ③ the fusion of the sensing results is not tampered with by the DSU, then one SU in this time slot will be able to obtain the right to access the spectrum and successfully transmit its own communication data packet.

[0078] The probability that the PU channel is idle is P0. The probability that m(t) HSUs participate in spectrum sensing in time slot t can be expressed by equation (18):

[0079]

[0080] Then, in a time slot, there will be an SU that can obtain the right to access the spectrum and successfully transmit its own communication data packet. The probability is expressed by formula (19):

[0081]

[0082] With probability P use (t) It is uncertain whether a data packet is successfully transmitted in a time slot, so P is used use (t) is compared with a probability threshold, which can be expressed by formula (20):

[0083]

[0084] When the probability P use When (t) is greater than the threshold probability, the data packet is transmitted successfully; when the probability P useWhen (t) is less than the threshold probability, the data packet transmission fails. In the tth time slot, after consensus, an SU wins the competition, and the SU successfully sends data, which is represented by s(t) = 1, and fails to send data, which is represented by s(t) = 0. Then whether the SU successfully sends data can be expressed by formula (21):

[0085]

[0086] The goal of this wireless network is to make full use of the idle spectrum resources of PU as much as possible, improve the utilization rate of spectrum resources, and thus improve network performance. a The larger the value, the more HSUs participate in sensing in a time slot, and the higher the accuracy of the fusion of sensing results to determine the busy and idle status of the PU. When the accuracy of spectrum sensing is very low, two situations may occur:

[0087] Case 1: The SU spectrum sensing fusion result recognizes the PU spectrum itself as an idle channel because it is in an occupied state. At this time, even if an SU obtains the right to use an idle channel based on the consensus mechanism, the communication information transmission in this time slot will fail due to the interference caused by the PU using the channel, and it will also cause interference to the PU, and the spectrum sharing will be declared a failure.

[0088] Case 2: The SU spectrum sensing fusion result recognizes the PU spectrum itself as idle as the channel is occupied. At this time, the SU as a whole will not continue the consensus process, so there is no SU that can use the PU spectrum resources for data communication, which reduces the efficiency of spectrum resource utilization. At this time, if the purpose of the DSU is to deceive other SUs to monopolize the idle channel, the DSU will transmit its own communication information in the channel stolen in this time slot, which greatly undermines the fairness of the network.

[0089] It can be seen that the accuracy of spectrum sensing is directly related to the quality of network performance. Therefore, a parameter is introduced: the perception accuracy γ as an indicator to measure network performance. It is defined as the ratio of the number of successful data transmissions by the SU that wins the consensus in all time slots to the number of times the cooperative spectrum sensing judges that the PU is idle, expressed by formula (22):

[0090]

[0091] Among them, T is the total number of time slots, and a γ value of 0.9 or above indicates good network performance. The larger the γ value, the higher the perception accuracy. The design of the lightweight consensus mechanism based on perception accuracy is based on perception accuracy. The more accurate the perception, the greater the chance of the SU consensus winning.

[0092] If all HSUs participate in spectrum sensing in all time slots, the accuracy of judging whether the PU is idle after the sensing results are fused will be the highest, so the idle spectrum utilization in the network will likely reach the maximum and the network performance will be optimal; however, since the i-th SU participating in spectrum sensing in the t-th time slot consumes additional energy c compared to when it does not sense. i (t), all HSUs participating in spectrum sensing in all time slots will cause the SU to generate the highest additional energy consumption. Therefore, the accurate spectrum sensing mechanism in an untrusted environment optimizes the probability value P of HSU participating in spectrum detection. a , achieving a compromise between the additional energy consumption of spectrum sensing and the network performance guided by the sensing accuracy, and enabling users to accurately judge the busy and idle status of the channel. The optimization problem can be expressed as formula (23):

[0093]

[0094] stc i (t)≤const (23-b)

[0095]

[0096]

[0097] Among them, const is the upper limit of additional energy consumption.

[0098] A genetic algorithm is used to solve the optimization problem to optimize the probability P of HSU participating in spectrum sensing. a , the pseudo code of the algorithm is described as follows:

[0099] pop: initial population

[0100] p c : Crossover probability

[0101] p b : Mutation probability

[0102] f: fitness

[0103] m: population size

[0104] g: Genetic Algebra

[0105]

[0106] 3. Lightweight consensus mechanism based on perceived accuracy

[0107] In a time slot, when the PU channel is idle, in order to select the SU that can use the idle spectrum to transmit data packets, it is hoped that a SU with a perception value closest to the cooperative spectrum perception result and a need for data transmission will be selected to obtain the spectrum use right first, because this SU is likely to be honest. At the same time, there are some HSUs that want to use the frequency band in every time slot and do their best to perform spectrum perception, but due to their own conditions, the accuracy of each perception is not the closest, so the integral value is introduced. The integral value takes into account the accuracy of perception and gives all HSUs the opportunity to use the channel.

[0108] Since DSU may release false information to reduce the efficiency of spectrum resource utilization, disrupt the normal operation of the network, or deceive other SUs to monopolize idle channels, the reputation value parameter is introduced to reduce the impact of DSU on the normal operation of the network.

[0109] The integral value and reputation value are obtained according to the scoring curve. The scoring curve must satisfy the difference between the spectrum sensing value of each SU participating in spectrum sensing and the fusion value of the spectrum sensing result x = |SNR(t)-SNR i The smaller (t)|, the more scores there are, that is, the score curve shows a monotonically decreasing trend. According to the golden section method, in a time slot, the scores between most SUs with smaller difference x after segmentation decrease slowly, and the score gap between a small number of SUs with larger difference x after segmentation is large, that is, the curve is similar to an S-shape, such as Figure 3 Typical S-shaped curves include Gompertz curve and Logistic curve. Polynomial curves can also be S-shaped and meet the scoring rules. The scoring curve is selected according to the specific network. For example, in a network with 100 SUs, the score of the i-th SU in the t-th time slot is obtained according to the polynomial curve: δ i (t) = x1x 5 +x2x 4 +x3x 3 +x4x 2 +x5x+x6, x1, x2, x3, x4, x5, x6 are polynomial curve parameters. In order to make the polynomial curve S-shaped and satisfy the scoring rule, let its values ​​be x1=-0.06195, x2=1.391, x3=-10.16, x4=25, x5=26.95, x6=100.

[0110] After the scores are accumulated over multiple time slots, the integral value and reputation value of the SU will continue to increase. However, the unlimited growth will lead to the possibility of borrowing PU idle channels converging on some SUs, which will also make the system unstable. Therefore, we use a cumulative upper limit v m , the cumulative value will increase infinitely and tend to v m , but will not reach v mThe accumulation of integral value and reputation value is related to the cumulative value at the beginning of a time slot according to the score curve. That is, under the same perception situation, the smaller the cumulative value, the more points are obtained after a time slot; the closer the cumulative value is to v m , the fewer the scores after a time slot, the overall cumulative curve is as follows Figure 4 As shown. Then the integral value and credit value obtained by the i-th SU in the t-th time slot can be expressed by equation (24) and equation (25):

[0111]

[0112]

[0113] The integral value of the i-th SU is expressed as The reputation value is expressed as In the initial state, there are Before the start of time slot 1, that is, in time slot 0, The reputation value is used as the confidence of the node, that is, the weight of the spectrum sensing result of the node during fusion, which is lower than the basic reputation value. The node will be considered as a DSU and lose the opportunity to use the idle shared spectrum resources. The integral value is used as the basis for using the shared channel to send data. Specifically, the integral value of the SU that wins the consensus mechanism and successfully occupies the channel to send data in each time slot is reduced by a sending cost. In order to monitor that the SU will not cooperate with the SU of the data receiving party to lie that the data packet was not successfully sent in order to avoid deducting the sending cost, while selecting the SU that can use the idle spectrum resources of the PU, a supervisory node is randomly selected from the remaining SUs. The supervisory node needs to sense whether there is communication data being sent during the spectrum use phase of a time slot, so as to ensure that data transmission failure can only be caused by the cooperative spectrum sensing mistakenly judging the busy state of the PU as an idle state. In the tth time slot, the ideological steps of the consensus mechanism are as follows:

[0114] (1) Each SU broadcasts whether it has data transmission requirements through the control channel. If the i-th SU participates in spectrum sensing, it also broadcasts its own spectrum sensing value SNR i (t). Assuming that the control channel is a reliable channel, the sensed value can be transmitted to all other SUs.

[0115] (2) Considering that all SUs can receive each other's spectrum sensing values, the integrated spectrum sensing values ​​calculated by all SUs are equal. For example, the i-th SU calculates the spectrum sensing value SNR based on its own spectrum sensing result and the spectrum sensing value SNR collected from other SUs. j (t)(j∈{1,2,…,n},j≠i), the collaborative spectrum sensing results of the whole network can be calculated

[0116] (3) The i-th SU calculates the difference between its own spectrum sensing value and the fusion value of the spectrum sensing result x = |SNR(t)-SNR i (t)|. Based on the difference, calculate the integral value you get in this time slot according to the scoring curve. Reputation Thus, the integral value is updated Update reputation

[0117] (4) The i-th SU broadcasts its updated integral value through the control channel Reputation

[0118] (5) Compare and judge whether the channel is idle. When the spectrum sensing result fusion value SNR(t) is greater than the threshold value ε(t), the channel is not idle, and no SU obtains access rights in this time slot, and enter (7); otherwise, the channel is idle, and enter (6).

[0119] (6) The SU with the highest integral value in this time slot (assuming it is the kth SU) obtains the right to generate blocks and use channels in this time slot, and randomly selects a supervisory SU. Therefore, the kth SU will use the idle channel in this time slot to send the data it needs to send. After the data is successfully sent, its integral value will be subtracted from the sending cost, that is,

[0120] (7) The SU with the highest integral value in this time slot will form a block for this time slot based on the fusion value of the spectrum sensing result of this time slot, the spectrum sensing value of each SU, the final integral value of each SU in this time slot and the reputation value, and write it into the blockchain.

[0121] (8) When the reputation value of a certain SU (assuming it is the dth SU) is lower than When a wireless router is in the same channel as the shared channel, it will be regarded as a DSU and lose the opportunity to compete for the idle shared channel.

[0122] Sending price According to the characteristics of the score curve and the integral value accumulation rules, the initial integral value For example, the slope of the integral value of 80-100 on the polynomial score curve is significantly smaller than the slope of the integral value between 0-80, that is, when the integral value is greater than 80, the change in the integral value of the SU participating in spectrum sensing after a time slot is small. In order to give all SUs an equal opportunity to use the PU idle channel, the increase in the integral value of the SU that has used the idle channel should be slower. At the same time, the SU that can obtain the right to use the PU idle channel according to the integral accumulation rule should not return to the initial competition state where it is impossible to distinguish whether it is an HSU or a DSU, so the cost of sending The value is 20.

[0123] Basic reputation value The initial reputation value is determined based on the characteristics of the score curve and the reputation value accumulation rules. For example, the slope of the polynomial score curve between -35 and -50 is smaller than that of the reputation value less than -50, which gives SUs with a reputation value between -35 and -50 room to maneuver. SUs can gradually increase their reputation values ​​by honestly publishing accurate spectrum sensing information. In this case, the reputation value is less than or equal to -50 because of the accumulated dishonest behavior of the SU, so the basic reputation value is The value is -50.

[0124] The lightweight consensus mechanism based on perception accuracy can fairly and reasonably select SUs that can use PU idle channels, while minimizing the impact of dishonest behavior of DSUs, and ensuring the efficiency and credibility of SUs in using idle channels in untrusted environments. Through the consensus algorithm based on spectrum perception accuracy, the scoring rules of integral value and reputation value are used. The integral value selects the SU that can use the idle spectrum in the way with the highest integral value in a time slot. The reputation value is used to adjust the weight of each SU in spectrum perception fusion to reduce the interference of DSU. The consensus algorithm is described as follows:

[0125]

[0126]

[0127] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and should be understood that the scope of protection of the present invention is not limited to such specific statements and embodiments. For those skilled in the art, the present invention may have various changes and variations. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of the claims of the present invention.

Claims

1. A spectrum sensing and trusted sharing method based on blockchain, characterized in that: The application scenarios include: primary user PU and secondary user SU. In an untrusted wireless network environment, each secondary user SU competes for the idle channel of the primary user PU. The untrusted wireless network environment is specifically aimed at the behavior that the spectrum sensing capabilities of each node are uneven and the nodes may covertly publish misleading spectrum sensing results due to their own communication needs. The specific sharing method includes the following steps: S1. Each secondary user SU broadcasts whether it has data transmission requirements through the control channel. If the i-th secondary user SU participates in spectrum sensing, it also broadcasts its own spectrum sensing value SNR i (t), t represents the tth time slot; S2. All secondary users obtain the collaborative spectrum sensing result SNR(t) of the entire network based on their own spectrum sensing values ​​and the spectrum sensing values ​​received from other secondary users; S3. The integral value and reputation value of each secondary user are updated as follows: the integral value and reputation value are obtained according to the score curve, and the score curve satisfies the difference between the spectrum sensing value of each SU participating in spectrum sensing and the fusion value of the spectrum sensing result x=|SNR(t)-SNR i The smaller (t)| is, the more points are scored, that is, the score curve shows a monotonically decreasing trend; and the updated integral value and reputation value are broadcasted through the control channel; The reputation value is used as the confidence of the node. The node with a reputation value lower than the basic reputation value is considered to be a dishonest secondary user DSU and loses the opportunity to use the idle shared spectrum resources; S4. If there is a secondary user SU whose credit value is less than the basic credit value, the secondary user SU loses the qualification to apply for an idle channel in this round; S5. If the cooperative spectrum sensing result of the entire network in this time slot is greater than the threshold value for judging whether the PU channel is busy or idle, it is considered that the current channel of the primary user PU is not idle, and step S7 is executed; otherwise, step S6 is executed; S6, the secondary user SU with the highest integral value in this time slot obtains the right to generate a block in this time slot and use the idle channel of the primary user PU. The secondary user SU uses the idle channel of the primary user PU in this time slot to send the data to be sent. After the data is successfully sent, the integral value of this time slot is deducted from the sending cost to obtain its integral value of this time slot; return to step S1 to compete for the idle channel of the primary user PU in the next time slot; S7. The secondary user SU with the highest integral value in this time slot will form a block for this time slot based on the collaborative spectrum sensing result of the entire network in this time slot, the spectrum sensing value of each secondary user SU, and the latest integral value and reputation value of each secondary user SU in this time slot, and write it into the blockchain; return to step S1 to compete for the idle channel of the primary user PU in the next time slot.

2. According to a blockchain-based spectrum sensing and trusted sharing method according to claim 1, it is characterized in that: Step S2 also includes: in a time slot, when the secondary user SU performs spectrum sensing result fusion, the dishonest secondary user SU will decide whether to tamper with its own spectrum sensing value in the current time slot according to its own spectrum sensing value of the primary user PU.

3. According to a blockchain-based spectrum sensing and trusted sharing method according to claim 2, it is characterized in that: The implementation process of dishonest secondary user SU tampering with its own spectrum sensing value is as follows: when ε i (t)-SNR i When (t)<d, the dishonest secondary user SU increases its spectrum perception value to So that the spectrum sensing results of the whole network are integrated to be greater than the threshold value for judging whether the PU channel is busy or idle; when SNR i (t)-ε i When (t)<d, the dishonest secondary user SU reduces its own spectrum perception value to So that the spectrum sensing results of the whole network are integrated and smaller than the threshold value for judging whether the PU channel is busy or idle; Among them, ε i (t) is the threshold value of spectrum sensing of the i-th secondary user SU in the t-th time slot, SNR i (t) represents the signal-to-noise ratio of the primary user PU signal perceived by the i-th secondary user SU in the t-th time slot, is the signal-to-noise ratio of the primary user PU signal perceived by the i-th dishonest secondary user SU in the t-th time slot, and d is a parameter.

4. According to claim 3, a spectrum sensing and trusted sharing method based on blockchain is characterized in that: The probability that a dishonest secondary user SU fails to tamper with its own spectrum sensing value is P r , and its calculation formula is: Among them, H(t)=1 means that the perception result fusion of all honest SUs participating in spectrum sensing in the tth time slot determines that the PU channel is idle, H(t)=0 means that the perception result fusion of all honest SUs participating in spectrum sensing in the tth time slot determines that the PU channel is busy, D(t)=1 means that the perception result fusion with dishonest SUs participating in the tth time slot determines that the PU channel is idle, and D(t)=0 means that the perception result fusion with dishonest SUs participating in the tth time slot determines that the PU channel is busy; H(t), D(t) represent the situation where dishonest SUs perform dishonest behavior; H(t)=1, D(t)=1 means that the perception result fusion of all honest SUs participating in spectrum sensing in the tth time slot determines that the PU channel is idle, and the perception result fusion with dishonest SUs participating in the tth time slot determines that the PU channel is idle; H(t)=0, D(t)=0 means that the perception result fusion of all honest SUs participating in spectrum sensing in the tth time slot determines that the PU channel is busy, and the perception result fusion with dishonest SUs participating in the tth time slot determines that the PU channel is busy.

5. According to a blockchain-based spectrum sensing and trusted sharing method according to claim 4, it is characterized in that: All honest SU with P a The probability of choosing whether to participate in spectrum sensing in this time slot is P a The value of is obtained by solving the following optimization problem: Among them, c i (t) indicates the extra energy consumed by the ith SU participating in spectrum sensing in the tth time slot compared to when it does not sense; const is the upper limit of the extra energy consumption; γ indicates the sensing accuracy; e(t) = 1 indicates that the sensing result fusion judges that the PU is idle; e(t) = 0 indicates that the sensing result fusion judges that the PU is busy; s(t) = 1 indicates that in the tth time slot, after consensus, an SU wins the competition and the SU successfully sends data; s(t) = 0 indicates that in the tth time slot, after consensus, an SU wins the competition and the SU fails to send data; ε i (t) is the threshold value of energy perception of the i-th SU in the t-th time slot, SNR(t) represents the signal-to-noise ratio of the PU signal after fusion, ε(t) represents the threshold value of energy perception after fusion, and the probability P is used use (t) measures whether a data packet is successfully transmitted in a time slot, which is uncertain. use (t) represents the probability that SU successfully sends data, Represents the probability threshold.

6. The spectrum sensing and trusted sharing method based on blockchain according to claim 5 is characterized in that: Step S3 updates the respective points and reputation values, specifically: For the i-th secondary user SU, calculate the integral value obtained in this time slot and reputation back, The update formula of the SU score value of the i-th user is: The update formula of the reputation value of the i-th secondary user SU is:

7. The spectrum sensing and trusted sharing method based on blockchain according to claim 6 is characterized in that: For the i-th secondary user SU, its own spectrum perception value is calculated as follows: in, They represent the channel fading power, PU signal power and channel noise power perceived by the i-th SU in the t-th time slot respectively.

8. The spectrum sensing and trusted sharing method based on blockchain according to claim 7 is characterized in that: The calculation formula for the collaborative spectrum sensing result of the entire network is: Among them, ω i (t) is the weight of the i-th secondary user SU in the t-th time slot when the sensing results are fused, and N(t) is the number of secondary user SUs participating in spectrum sensing in the t-th time slot.

9. The spectrum sensing and trusted sharing method based on blockchain according to claim 8, characterized in that: ω i The calculation formula of (t) is:

Images