A cognitive internet of things interference management method based on interference steering

Abstract

This invention relates to a cognitive IoT interference management method based on interference redirection, comprising: designing precoding and decoding vectors according to the CSI between PU-Tx and PU; determining whether the current CU is in its corresponding working time slot; and determining whether the CU is in the working time slot. n Design precoding and decoding vectors; use TDD-EI-SCIS to eliminate interference signals in cognitive IoT; determine whether the power overhead of the steering signal to be transmitted by PU-Tx meets the transmission conditions; if the conditions are met, use TDD-EI-SCIS to manage PU and CU. n The system detects interference signals from CU-Tx and PU-Tx at each location; it uses TDD-SCIS to design the precoding vector and transmit power overhead for the corresponding steering signals of multiple interference signals; CU-Tx and PU-Tx simultaneously transmit their respective steering signals and desired signals. This invention effectively eliminates interference between secondary users.

Description

TECHNICAL FIELD

[0001] The present application relates to a kind of joint management method of equivalent interference subchannel interference diversion (TDD-EI-SCIS) based on time division duplex (TDD), TDD-based subchannel interference diversion (TDD-SCIS) and partial interference alignment (PIA) in cognitive internet of things (Cognitive IoT, C-IoT) Multi-user Multi-data flow scene, specifically, it is a kind of interference diversion based cognitive internet of things interference management method. BACKGROUND

[0002] C-IoT is considered as a promising solution to solve the spectrum resource scarcity by integrating human cognitive process into current Internet of Things (IoT). Through dynamic spectrum access (DSA) technology, C-IoT devices (C-IoT Devices, C-IoD) are allowed to share licensed spectrum with authorized users (also known as primary users (Primary user, PU)) as unlicensed users (also known as cognitive users (Cognitive user, CU), or secondary users (Secondary user, SU)). According to the different coexistence modes of PU and CU, the DSA model can be divided into opportunity spectrum access (OSA) model and concurrent spectrum access (CSA) model. Under the OSA model, CU needs to perform dynamic spectrum control, which brings a burden to low-cost IoT devices. In order to alleviate this burden, the CSA model is proposed, in which the CU transmitter (CU-Tx) needs to control the transmission power so that the interference caused to the PU satisfies the interference temperature condition, while the required signal received by the CU from the CU-Tx is often overwhelmed by various interference signals. People introduce interference management in C-IoT to make CU and PU meet the basic conditions of spectrum sharing, that is, CU will not have unacceptable impact on the quality of service (QoS) of PU during accessing the spectrum.

[0003] At present, the existing interference elimination method in C-IoT is mainly IA, and these methods mainly eliminate the interference suffered by PU, ignoring the performance requirements at CU and the interference management in cognitive network, resulting in the decline of network capacity of cognitive network and the QoS at CU cannot be guaranteed, and IA is highly dependent on system configuration parameters, and its applicability is often limited by the degree of freedom. SUMMARY

[0004] The purpose of this invention is to provide a cognitive IoT interference management method based on interference redirection. It combines advanced IS technology with IA to eliminate interference in C-IoT to improve licensed spectrum utilization, and improves the communication quality of CU as much as possible while ensuring that the QoS of PU is not affected.

[0005] The technical solution adopted by this invention to solve its technical problem is as follows: This cognitive IoT interference management method based on interference redirection is a joint management method for interference management in cognitive IoT, including the following steps:

[0006] Step 1: PU-Tx and CU-Tx share Channel State Information (CSI) and data information, wherein the number of transmit antennas in PU-Tx is N. T,p The number of transmit antennas in CU-Tx is N. T,c The number of receiving antennas for the PU is N. R,p The nth CU is CU n CU n The number of receiving antennas is N R,n And satisfy N T,p ≥N T,c ≥N R,i Where i = n, p, and the desired signal received by PU is x. p =[x p (1),...,x p (k p ),...,x p (K p )] T , where x p (k p ),(k p ∈{1,2,...,K p},K p ≤N R,p ) is the kth number received by PU p One data symbol, CU n The received desired signal is x n =[x n (1),...,x n (k n ),...,x n (K n )] T , where x n (k n ),(k n ∈{1,2,...,K n},K n ≤N R,n ) is the kth number received by PU n One data symbol;

[0007] Step two, design the precoding vector and decoding vector of each data symbol according to the CSI between PU-Tx and PU by constructing Hermitian semi-definite matrix;

[0008] Step three, divide the corresponding working time slot for each CU in the cognitive network, assuming the number of CUs is N, the total time is T, the total number of time slots in the cognitive network is N, and each time slot occupies T / N time. For the nth CU, i.e. CU n , when the nth time slot is the working time slot of CU n , the remaining N-1 time slots are all its interference time slots. At this time, design the precoding vector and decoding vector of each data symbol according to the CSI between CU-Tx and CU by constructing Hermitian semi-definite matrix. Repeat the operation of steps four to eleven for the CUs in the other N-1 working time slots of the cognitive network;

[0009] Step four, PU and CU n are subjected to multiple interference signals from CU-Tx and primary transmitter (PU-Tx) respectively. First, use TDD-EI-SCIS for interference management, and equivalently regard the multiple interference signals received by each as an effective interference signal;

[0010] Step five, CU-Tx and PU-Tx design the corresponding steering signals and respectively for the equivalent interference signals, and the precoding vector and transmission power overhead of the steering signal transmitted by CU-Tx are and respectively, and the precoding vector and transmission power overhead of the steering signal transmitted by PU-Tx are and

[0011] Step six, judge whether the transmission condition of PU-Tx is met. If it is met, use TDD-EI-SCIS to manage the interference signals from CU-Tx and PU-Tx respectively at PU and CU n ; if it is not met, CU n still uses TDD-EI-SCIS to manage the interference from PU-Tx, while the interference management at PU performs the operations of steps seven to ten;

[0012] Step seven, use TDD-SCIS for the multiple interference signals from CU-Tx received at PU, and the precoding vector and transmission power overhead of the steering signal required at PU-Tx for each interference signal are and respectively, where k n ∈{1,2,...,Kn};

[0013] Step eight, according to the transmission condition of PU-Tx, the data symbols corresponding to the combination satisfying the condition are set as set A, and vice versa as set B; n (k n ) is set as set A, and vice versa as set B;

[0014] Step nine, the interference caused by the data symbols in set A to PU in the transmission process is managed by TDD-SCIS;

[0015] Step ten, the interference signal corresponding to the data symbols in set B cannot be managed by TDD-SCIS, and needs to be aligned to the interference subspace designed at PU by PIA, at this time, CU-Tx needs to adjust the precoding vector corresponding to this part of data symbols;

[0016] Step eleven, CU-Tx and PU-Tx transmit their respective turning signals and expected signals, CU n and PU use their respective decoding vectors to recover the expected signals.

[0017] The step ten in the above scheme is specifically:

[0018] Assuming that the number of data symbols in B is Ω, the precoding matrix of this part of data symbols is n (δ), δ∈Ω Wherein, the value of δ is discontinuous, that is, Ω data symbols are extracted from the original K n data symbols; After PIA processing, it becomes The interference subspace designed at PU is

[0019] The precoding matrix under PIA needs to satisfy:

[0020]

[0021] Wherein, is the channel matrix from CAP to PU, so that The condition for having a solution is rank(G p ) = rank(G p |Δ) = N R,p , when N R,p = N T,c , When N R,p < N T,c , The minimum norm solution of is Wherein denotes G p is the pseudo-inverse of G should be N T,c ×(N T,p -K p ) matrix, when Ω≤N T,p -K p , the interference subspace at PU is large enough to satisfy the transmission of the Ω data symbols on their respective sub-channels; when Ω>N T,p -K p , Ω-(N T,p -K p ) data symbols need to share the sub-channels with the first N T,p -K p data symbols, so the column number of G needs to be modified to min{Ω,N T,p -K p};

[0022] The PIA constraint condition of the partial interference signal in B should satisfy,

[0023]

[0024] wherein, is a decoding matrix composed of the first K p decoded vectors f p (k p ) at PU, is a decoding matrix composed of the decoding vectors corresponding to the data symbols in B.

[0025] The present application has the following beneficial effects:

[0026] 1. The proposed TDD-EI-SCIS solves the drawbacks of the conventional IS technology (MTIS, STIS, AIS) which relies on the steering mode of the expected signal, i.e. in the multi-user multi-data flow scenario, the interference after steering will interfere with other data flows except the reference expected data flow of the CU, and when there are multiple CUs in the cognitive network, the conventional IS technology cannot be used to eliminate the interference between secondary users, while the TDD-EI-SCIS can.

[0027] 2. In view of the problem that the priority of the PU using the licensed spectrum is higher than that of the CU in C-IoT, a joint management method based on TDD-EI-SCIS, TDD-SCIS and PIA is further proposed on the basis of TDD-EI-SCIS, which solves the problem that TDD-EI-SCIS cannot eliminate interference at the PU in some scenarios.

[0028] 3、The present application combines advanced IS technology with IA to eliminate interference in C-IoT to improve licensed spectrum utilization, and under the premise of ensuring that the QoS of PUs is not affected, the communication quality of CUs is improved as much as possible. BRIEF DESCRIPTION OF DRAWINGS

[0029] Figure 1 is a flow chart of the method of the present application.

[0030] Figure 2 is a cognitive internet of things system model to which the method of the present application is applied.

[0031] Figure 3 is a graph of the relationship between the spectrum efficiency and the signal-to-noise ratio (SNR) of CUs in the cognitive network and PUs in the main network in the T time period according to the method of the present application. DETAILED DESCRIPTION

[0032] The present application is further described below in conjunction with the accompanying drawings:

[0033] In conjunction with Figure 1 the present application is a new IS method based on interference redirection, and is applied to C-IoT, and further proposes a joint management method based on TDD-EI-SCIS, TDD-SCIS and PIA, which specifically eliminates the interference suffered by PUs and CUs in C-IoT, and the specific steps are as follows:

[0034] 1. Design the precoding vector and the decoding vector according to the CSI between PU-Tx and PUs;

[0035] 2. Determine whether the CU n at this time is in its corresponding working time slot;

[0036] 3. For the CU n located in the working time slot, design the precoding vector and the decoding vector according to the CSI between CU-Tx and CUs n ;

[0037] 4. Use TDD-EI-SCIS to eliminate the interference signals in the cognitive internet of things, and PU-Tx and CU-Tx design the precoding vector and the transmission power overhead of the corresponding redirected signal according to the equivalent interference at PUs and CUs n ;

[0038] 5. Determine whether the power overhead of the redirected signal to be transmitted by PU-Tx meets the transmission condition, and if the condition is met, use TDD-EI-SCIS to manage the interference signals from CU-Tx and PU-Tx suffered by PUs and CUs n , respectively;

[0039] 6. When the power overhead of the steering signal that PU-Tx needs to transmit does not satisfy the transmission condition, for the case that PU is interfered by multiple interferers from CU-Tx, the processing method of combining TDD-SCIS and PIA is used, while the interference from PU-Tx to CU is still processed by TDD-EI-SCIS;

[0040] 7. The corresponding steering signal precoding vector and transmission power overhead of multiple interference signals are designed by using TDD-SCIS;

[0041] 8. The steering signals corresponding to the interference signals that satisfy the transmission condition of PU-Tx are judged and selected. The set of data symbols corresponding to the interference signals that satisfy the condition is A, and the set of data symbols corresponding to the interference signals that do not satisfy the condition is B.

[0042] 9. The interference of the data symbols in set A to PU is managed by using TDD-SCIS in the transmission process, while the data symbols in set B are adjusted by using PIA to align the interference of the data symbols to the interference subspace designed at PU.

[0043] 10. CU-Tx and PU-Tx transmit their respective steering signals and desired signals at the same time, and the desired signals are recovered at CU n and PU by using their respective decoding vectors.

[0044] Embodiment:

[0045] The system model referred to in this embodiment is shown in Figure 2 , considering that PU-Tx works in a residential area of 300m x 300m, the transmission power is about 30dBm, and the path loss under the coverage of PU-Tx is PL p (dB) = 140.7 + 36.7log 10 R(km). CU-Tx is located in a certain house, and its radio coverage does not exceed 50m, the transmission power is 20dBm, and the path loss is PL n (dB) = 61.38 + 17.97log 10 r(m). The value range of is -71dBm to 20dBm, The value range of is -91dBm to 30dBm. For the sake of clear expression, is used is also defined and It is assumed that the distance from PU-Tx to PU is equal to the distance from PU-Tx to CU n , the distance from CU-Tx to PU is equal to the distance from CU-Tx to CU n , that is, PL n = PLpn and PL p =PL np This ensures that the PU distribution is random while taking into account that the PU is located within the CU-Tx coverage area. Based on the above parameter settings, ξ1∈[-101,111]dB. Since CU-Tx cannot be deployed close to PU-Tx in actual scenarios, this invention sets ξ1∈[0.1,100] during simulation and performs simulation using the Monte Carlo simulation method.

[0046] Set ξ1 = 50, N = 3, K n =3,K p =3,N R,n =N T,c =N R,p =N T,p =6, there are three cognitive users in the cognitive network. CU-Tx transmits three desired data symbols to each CU, and PU-Tx transmits three desired data symbols to each PU. PU is subjected to three interference signals from CU-Tx in each time slot. n The system is subjected to three interference signals from PU-Tx during its operating time slot. Under the above simulation conditions, the method of this invention is used to determine the relationship curves between the spectral efficiency (SE) and signal-to-noise ratio (SNR) of the CU in the cognitive network and the PU in the main network during the time period T. The results are as follows. Figure 3 As shown, Figure 3 A joint management method based on TDD-EI-SCIS, TDD-SCIS, and PIA was compared with TDD-EI-SCIS alone. Under the simulation conditions, the SE of the PU in the main network was improved by about 43 bits / s / Hz in the joint management method compared with TDD-EI-SCIS, while the SE of the CU in the cognitive network was reduced by about 17 bits / s / Hz. The total SE of the entire system was improved from about 137 bits / s / Hz under TDD-EI-SCIS to about 163 bits / s / Hz under the joint management method. The results show that the joint management method not only improves the SE of the main network under TDD-EI-SCIS without having a significant impact on the cognitive network, but also significantly improves the SE from the perspective of the entire system.

[0047] This invention performs precoding and power control at the primary base station in the main network, and decoding at the primary and secondary users in the cognitive network. It also incorporates TDD technology to eliminate interference from secondary users to primary users and interference between secondary users.

[0048] The above is the preferred embodiment of the present application, but the present application should not be limited to the embodiment and the disclosure of the drawings, and the method has important application value in cognitive internet of things interference management. Therefore, any equivalent or modification completed without departing from the disclosed technical method falls within the scope of the present application.

Claims

1. A cognitive Internet of Things interference management method based on interference steering, characterized in that The method comprises the following steps: Step 1: The main transmitter PU-Tx and the CU transmitter CU-Tx share Channel State Information (CSI) and data information. The number of transmit antennas in the PU-Tx is N. T,p The number of transmit antennas in CU-Tx is N. T,c The number of receiving antennas for the primary user PU is N. R,p The nth cognitive user CU is CU n CU n The number of receiving antennas is N R,n And satisfy N T,p ≥N T,c ≥N R,i Where i = n, p, and the desired signal received by PU is x. p =[x p (1),...,x p (k p ),...,x p (K p )] T , where x p (k p ),(k p ∈{1,2,...,K p },K p ≤N R,p ) is the kth number received by PU p One data symbol, CU n The received desired signal is x n =[x n (1),...,x n (k n ),...,x n (K n )] T , where x n (k n ),(k n ∈{1,2,...,K n },K n ≤N R,n ) is the kth number received by PU n One data symbol; Step two, designing the precoding vector and decoding vector of each data symbol by constructing Hermitian semi-definite matrix according to the CSI between PU-Tx and PU; Step three, divide the corresponding working time slot for each CU in the cognitive network, assuming the number of CUs is N, the total time is T, the total number of time slots in the cognitive network is N, each time slot occupies T / N time, for the nth CU, that is, CU n When the nth time slot is the working time slot of CU n , the remaining N-1 time slots are all its interference time slots, at this time, according to the CSI between CU-Tx and CU, the pre-coding vector and the decoding vector of the respective data symbol are designed by constructing the Hermitian semi-definite matrix, and the operations of steps four to eleven are repeated for the CUs in the other N-1 working time slots of the cognitive network. Step four, for PU and CU n Subjected to multiple interfering signals from CU-Tx and the main transmitter PU-Tx respectively, first adopt TDD-EI-SCIS to carry out interference management, and equivalent multiple interfering signals subjected respectively to one effective interfering signal; Step five, CU-Tx and PU-Tx design corresponding steering signals respectively to equivalent interference signals and The pre-coding vector and transmit power overhead of the steering signal transmitted by the CU-Tx are respectively and The pre-coding vector and transmit power overhead of the steering signal transmitted by the PU-Tx are respectively and Step six, judging whether the transmission condition of PU-Tx is satisfied, if yes, using TDD-EI-SCIS to manage PU and CU n respectively suffered from the interference signals from CU-Tx and PU-Tx; if not, CU n still using TDD-EI-SCIS to manage the interference from PU-Tx, while the interference management at PU performs the operations of steps seven to ten; Step seven, using TDD-SCIS for the multiple interference signals suffered by the PU from the CU-Tx, each interference signal requiring a steering signal at the PU-Tx The precoding vector and the transmit power overhead of the steering signal at the PU-Tx are and where k n ∈{1,2,...,K n} Step eight, according to the transmission condition of PU-Tx, for the data symbol x corresponding to the combination of the PU-Tx satisfying the condition n (k n ) is set as set A, and vice versa as set B;​ Step nine, using TDD-SCIS to manage the interference caused by the data symbol in set A to PU in the transmission process; Step ten, the interference signal corresponding to the data symbol in set B cannot be managed by TDD-SCIS, and needs to be aligned to the interference subspace designed at PU by using PIA, at this time, the CU-Tx needs to adjust the precoding vector corresponding to the data symbol; Step eleven, CU-Tx and PU-Tx transmit their respective steering signals and desired signals simultaneously, CU n and PU recover the desired signals using their respective decoding vectors.

2. The interference steering based cognitive internet of things interference management method according to claim 1, characterized in that: The step ten is specifically: Assume the number of data symbols in B is Ω, the precoding vector z of this part of data symbols is n The precoding matrix composed of (δ), δ∈Ω is Wherein, the value of δ is discontinuous, from the original K n data symbols, δ data symbols are extracted; After PIA processing, it becomes The interference subspace designed at PU is Precoding matrix under PIA The following needs to be satisfied: in, Let be the channel matrix from CAP to PU, such that The condition for a solution is rank(G) p ) = rank(G p |Δ)=N R，p When N R,p =N T,c hour, When N R,p <N T,c hour, The minimal norm solution is in G represents p The false rebellion; at this time, It should be N T,c ×(N T,p -K p A 3D matrix, when Ω≤N T,p -K p When Ω > N, the interference subspace at PU is large enough to satisfy the transmission of these Ω data symbols on their respective sub-channels; T,p -K p At that time, Ω-(N T,p -K p ) data symbols need to be combined with the first N T,p -K p Each data symbol shares a sub-channel, so The number of columns needs to be modified to min{Ω,N T,p -K p }; The PIA constraint condition of the part of interference signal in B should satisfy, wherein, is a decoding matrix composed of the first K p decoded vectors f p (k p ) at the PU, is a decoding matrix composed of the decoded vectors corresponding to the data symbols in B.

Images