WO2020206604A1 - 一种毫米波通信方法和通信系统 - Google Patents

一种毫米波通信方法和通信系统 Download PDF

Info

Publication number
WO2020206604A1
WO2020206604A1 PCT/CN2019/081914 CN2019081914W WO2020206604A1 WO 2020206604 A1 WO2020206604 A1 WO 2020206604A1 CN 2019081914 W CN2019081914 W CN 2019081914W WO 2020206604 A1 WO2020206604 A1 WO 2020206604A1
Authority
WO
WIPO (PCT)
Prior art keywords
beam width
user
carrier
price
millimeter wave
Prior art date
Application number
PCT/CN2019/081914
Other languages
English (en)
French (fr)
Inventor
伍楷舜
龙金凤
王璐
华慧丰
Original Assignee
深圳大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳大学 filed Critical 深圳大学
Priority to PCT/CN2019/081914 priority Critical patent/WO2020206604A1/zh
Publication of WO2020206604A1 publication Critical patent/WO2020206604A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W52/00—Power management, e.g. TPC [Transmission Power Control], power saving or power classes
    • H04W52/04—TPC
    • H04W52/30—TPC using constraints in the total amount of available transmission power
    • H04W52/34—TPC management, i.e. sharing limited amount of power among users or channels or data types, e.g. cell loading
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W72/00—Local resource management
    • H04W72/04—Wireless resource allocation

Definitions

  • the present invention relates to the field of wireless communication technology, and in particular to a millimeter wave communication method and communication system.
  • millimeter wave with its unique bandwidth advantage and unlicensed spectrum can solve the problem of "bandwidth hunger" and provide the communication network with gigabit-level communication requirements and ultra-low latency transmission.
  • millimeter wave networks are not a simple upgrade of low-frequency networks.
  • Friis transmission equation Francesian transmission equation
  • the loss of the free space propagation path is proportional to the second power of the frequency.
  • millimeter waves face huge transmission loss.
  • millimeter wave uses beamforming technology to generate directional gain to compensate for this transmission loss.
  • the problems of such directional narrow beam transmission are: (1) Millimeter wave transceivers that use directional transmission need to be established effectively Corresponding beam scanning and alignment are performed before the link, but the time complexity and space complexity of this process are relatively high. Once the link changes (for example, the location of the receiver changes), the original communication link will be unusable.
  • the research and application of millimeter wave communications are mainly focused on point-to-point links.
  • the use of spatial freedom is another way to meet the increasing data rate and service quality Methods. For example, by deploying multiple antennas at the transmitting end and the receiving end to form a multiple-input multiple-output (MIMO) system, this transmission method achieves spatial diversity, multiplexing gain and increased capacity.
  • MIMO multiple-input multiple-output
  • a massive array antenna to generate multiple narrow beams can improve beam multiplexing gain and reduce inter-user interference , Enabling access devices (such as base stations, access points, etc.) to further utilize spatial freedom to serve multiple mobile users.
  • Enabling access devices such as base stations, access points, etc.
  • the high throughput of a large-scale multiple-input multiple-output system depends on the channel state information of the base station. In a frequency division multiplexing system, the feedback of channel state information is expensive due to the large-scale antenna.
  • BDMA wavelength division multiplexing
  • millimeter wave communication networks can use beamforming, multiple input multiple output (MIMO) and wavelength division multiplexing technologies to improve network performance
  • MIMO multiple input multiple output
  • wavelength division multiplexing technologies to improve network performance
  • the purpose of the present invention is to overcome the above-mentioned defects of the prior art and provide a millimeter wave communication method and communication system.
  • a millimeter wave communication method includes the following steps:
  • Step S1 For millimeter wave transmission, determine the optimized beam width and the user corresponding to the optimized beam width with throughput as the optimization target;
  • Step S2 For multiple users corresponding to the optimized beam width, concurrent transmission is performed using multiple access based on constellation redundancy.
  • step S1 includes the following sub-steps:
  • step S2 includes the following sub-steps:
  • the code bits are mapped to the constellation symbol set to obtain orthogonal codes.
  • the hash operation includes:
  • the user message is divided into multiple data blocks, a hash operation is performed on each data block, and the result of the hash operation performed on the previous data block is used as the input of the subsequent data block to perform the hash operation.
  • step S2 for multiple users corresponding to the optimized beam width, carrier allocation is performed based on the auction theory.
  • step S2 includes the following sub-steps:
  • the user receives the allocatable carrier and predicts the price of the allocatable carrier based on the historical observation value of the auction, and obtains the predicted price of the allocatable carrier;
  • the user allocates power to the allocatable carrier according to the predicted price and channel quality of the allocatable carrier;
  • the user determines the carrier and bid price to be auctioned
  • the user receives the bidding result, where the bidding result includes the winner's price.
  • the price prediction of the allocatable carrier includes:
  • Each user records the highest auction history price of each carrier as an observation value, and predicts the price of the next time slot subcarrier based on the observation value to determine the bid price.
  • a millimeter wave communication system includes:
  • the optimized beam width and the user corresponding to the optimized beam width are determined with throughput as the optimization target.
  • the access device is configured to perform: setting a minimum beam width, a maximum beam width, and a beam width change amount; taking the beam width change amount as a step size, the minimum beam width and the maximum beam width Scan between the beam widths to find the beam width corresponding to the maximum throughput of the system as the optimized beam width.
  • the present invention has the advantages of: for millimeter wave transmission, variable beam width is used to stimulate more concurrent transmissions, rateless codes are used to adapt to channel changes, and new multiple access technologies are used for concurrent transmission. transmission.
  • the auction theory is used to strategically allocate spectrum resources and the user's own power, thereby improving the user's transmission efficiency, and solving the problem of low spatial reuse capacity caused by the highly directional transmission of millimeter wave communications.
  • Fig. 1 shows a model diagram of a millimeter wave communication system according to an embodiment of the present invention
  • Fig. 2 shows a flowchart of a method for adaptive adjustment of beam width according to an embodiment of the present invention
  • Figure 3 shows a multiple access architecture based on constellation redundancy according to an embodiment of the present invention
  • Figure 4 shows a schematic diagram of an encoding and decoding process according to an embodiment of the present invention
  • Fig. 5 shows a flowchart of a spectrum allocation method based on auction theory according to an embodiment of the present invention.
  • the embodiments provided by the present invention can not only allow multiple users to access in a millimeter wave communication network, but also ensure reliable transmission according to channel quality, thereby improving spatial multiplexing efficiency.
  • the embodiments of the present invention also consider the problems of spectrum allocation and power allocation. The following will focus on embodiments of beam width adjustment, multiple access coding and decoding, spectrum allocation, and power allocation.
  • the embodiment of the present invention adjusts the beam width adaptively. For example, as shown in Fig. 1, for the case of one AP (Access Point) and multiple users, users are randomly distributed near the AP, where the AP generates multiple directional beams through array antenna units to serve the users. In a dense user scenario, one beam may cover multiple users. Assuming that the AP can communicate with multiple users at the same time in the beam, the relationship between the directional gain of the beam and the beam width can be known through the ideal sector antenna model, that is, the wider the beam width, the smaller the directional gain of the beam; on the contrary, the greater the beam width Narrow, the greater the directional gain of the beam.
  • the directional gain of the beam can improve the user’s signal-to-noise ratio.
  • Shannon’s formula the greater the signal-to-noise ratio, the greater the transmission rate. Therefore, the greater the beam directional gain, the greater the user’s transmission rate and the narrower the beam width. The narrower the width, the fewer users covered.
  • increasing the beam width can increase the number of users covered, thereby increasing the system throughput, but the transmission rate of a single user cannot be guaranteed.
  • reducing the beam width can increase the transmission rate of a single user, thereby increasing the system throughput, but the number of users covered may be reduced. Based on the above principle, there is a maximum throughput, and the beam width when the maximum throughput is achieved is called the optimal beam width in the present invention.
  • the maximum system throughput is taken as the optimization goal to find the optimal beam width. For example, starting from the smallest beam width, increase the beam width in sequence, calculate the system throughput each time after increasing the beam width, and compare it with the system throughput of the previous calculation, and take the beam width corresponding to the maximum system throughput as the maximum Optimal beam width.
  • the beamwidth adaptive adjustment method includes the following steps:
  • Step S210 Set the minimum beam width and calculate the corresponding throughput.
  • the minimum beam width and the maximum beam width can be set according to the scale of the antenna array, the user's demand for transmission quality, and the user density. Further, the system throughput corresponding to the minimum beam width is calculated.
  • Step S220 Increase the beam width and calculate the corresponding throughput.
  • the beam width is increased by a certain amount each time, and the corresponding system throughput is calculated.
  • the number of beam width increases each time can be set according to the execution speed and beam width resolution, for example, if you need to quickly determine the optimal For the beam width, a larger number of beam widths are increased each time, and if a more accurate optimal beam width needs to be determined, a smaller number of beam widths are increased each time.
  • Step S230 compare whether it is greater than the maximum beam width.
  • step S240 the system throughput corresponding to the current beam width is compared with the system throughput corresponding to the previous beam width. If it is greater than and the current beam width has reached the set maximum beam width, then step S240 is executed. If the current beam If the width does not reach the set maximum beam width, step S220 is continued.
  • Step S240 Determine the beam width and the user when the throughput is maximum.
  • the beam width corresponding to the maximum throughput of the system, record it as the optimal beam width and record the users covered at this time, which can be one user or multiple users .
  • the embodiment of the present invention provides a multiple access technology in which multiple users perform concurrent transmission.
  • This multiple access technology can adapt to channel state changes and achieve the following objectives:
  • the AP communicates with a group of users simultaneously , AP encodes all messages into a group of non-rate codes, and selects a transmission rate higher than the maximum rate allowed by the link conditions to broadcast.
  • the target user receives this group of non-rate codes
  • each user continues to extract from this group of codes
  • Useful information is output until its own message is successfully decoded, and then each user sends a feedback message to the AP, and the AP receives all the feedback messages before proceeding to the next round of transmission.
  • each user encodes their own data packets into a set of no-rate codes.
  • the present invention provides an improved encoder and decoder.
  • a hash-based encoder is used.
  • the encoder includes a sequential encoder, a random number generator (also known as a random symbol generator), and an orthogonal constructor.
  • the sequential encoder and the random number generator are used to maintain a robust transmission link
  • the orthogonal constructor is used to provide concurrent transmission in the constellation space.
  • This kind of encoder has the characteristics of randomness, orthogonality, linearity, etc. Randomness is the core of using dense constellation diagram transmission, and the flexibility to noise and interference will be significantly improved; Orthogonality is a structure that guarantees the separation of multiple users. It is very necessary to construct an orthogonal structure in the graph space; linearity is the key to a decoder with polynomial time complexity and space complexity, and it is also indispensable for a linear decoder in a multi-user scenario.
  • the constellation diagram usually represents the modulated signal in the vector space, and each signal has a position in the constellation diagram, called the constellation point.
  • Different modulation modes indicate the positions of all possible constellation points, and different modulation modes make the distance between constellation points different.
  • Wireless networks usually use fixed and symmetrical constellations, such as convolutional codes and low-density parity-check codes, but in the case of poor channel quality, the fixed distance of the constellation points will limit the rate and transmission rate.
  • the multiple access technology using the hash function in the embodiment of the present invention inherits several advantages of the rateless code.
  • the encoder uses a hash function to generate a random, non-linear mapping between message bits and coded bits, and then map code bits (code bits) to a dense constellation symbol set to transmit as much information as possible . Because of the characteristics of the hash function itself, the old status code and the new status code have the same length, and for any two input information, even if only one bit is different, the generated results are completely different. In this way, it can be effective Fight against noise and interference.
  • the millimeter wave multiple access technology of the embodiment of the present invention constructs an orthogonal coded bit for user separation.
  • a set of initial state values (the encoder at the transmitter and the decoder at the receiver know each other, for example, all values are all 0) are divided into several segments, and each segment is assigned to a user. Divide the user's message into several small blocks, and then execute the sequential hash function on the message block. Each time the hash function is executed, a result will be produced. Due to the nature of the hash function, these results can form an orthogonal space.
  • the initial state value is received, the initial state value zero is filled into the orthogonal space to supplement the length of each message block. In this way, users from concurrent transmissions are orthogonal in the constellation dimension, so that they can be decoded correctly.
  • the embodiment of the present invention uses a linear congruential generator (LCG) or a variant thereof to replace a non-linear random generator (RNG) for randomization, which ensures the performance of the decoder in a multi-user scenario. Practicality.
  • LCG linear congruential generator
  • RNG non-linear random generator
  • the decoding algorithm for millimeter wave multiple access in the embodiment of the present invention obtains advantages from the coding structure. For example, by searching a tree based on Maximum Likelihood (ML) to find the branch with the least cost as aims.
  • ML Maximum Likelihood
  • the key challenge is how to implement an ML decoder with guaranteed performance and minimal computational overhead. Due to the linear nature of the encoder, the receiver only needs to build its own decoding tree, and the decoding result is the least expensive branch.
  • an M-algorithm based on soft information is used to prune the tree, thereby further reducing the decoding complexity.
  • the maximum likelihood decoder has an exponential decoding complexity. With the increase in the number of multiple accesses, it is impossible to achieve pruning by simply using the M-algorithm, because sometimes it may be correct due to distortion.
  • the code is discarded, but the discarded code carries a large amount of reliable soft information. By extracting the soft information, the decoding accuracy can be improved and the calculation overhead can be reduced.
  • This soft information is called "Log-likelihood Ratio" (Log-likelihood Ratio, LLR), which indicates the certainty of the decoder for each bit decision and is used to correct the final decoding decision.
  • the number of concurrent transmissions that an N-order modulated constellation can support is N, because there are potentially N orthogonal constellation points on the constellation.
  • the modulation order increases, correct decoding requires better channel conditions. Therefore, the maximum number of concurrent transmissions that can be supported cannot exceed the Shannon limit.
  • the throughput of each user will decrease, so in fact, it is necessary to find a balance between concurrent transmission and the throughput of each user.
  • Fig. 3 shows a millimeter wave-based communication system according to an embodiment of the present invention.
  • the system includes a transmitter and a receiver.
  • the processing of the transmitter includes scrambling, encoder, orthogonal constructor, random symbol generator, and constellation diagram.
  • the mapper after constellation map mapping, the generated modulated signal is processed by the physical layer and sent to the receiving end via the RF front-end, corresponding to the transmitting end.
  • the processing of the receiving end includes constellation diagram mapping, multi-user detection, sequential decoder, and solution. Disturb etc.
  • Figure 3 illustrates a typical process of data transmission and reception in a communication network.
  • This process belongs to the prior art, such as scrambling, descrambling, CRC check, and retry count of the transmission process, etc., here No longer.
  • the contribution of the present invention is that the provided encoder includes a sequential encoder and a random symbol generator to maintain the robustness of the transmission link, and the orthogonal constructor provides concurrent transmission in the constellation space; the constellation decoder is based on soft information and maximum likelihood However, it is estimated to realize the actual decoder with the smallest decoding complexity.
  • the present invention adapts the link quality and user requirements through adaptive adjustment of the beam width.
  • the encoding and decoding process provided by the embodiment of the present invention is shown in FIG. 4.
  • the message M of User1 is divided into three different symbols m 1 , m 2 , and m 3 , each The symbol has k bits.
  • the initial state of user 1 and the first symbol m 1 are generated by a hash encoder (marked as h) to generate s 1
  • s 1 and m 2 are generated by a hash encoder to generate s 2.
  • This step is called sequential coding.
  • the initial state values are filled into s 1 , s 2, and s 3 to generate s 1 , s 2 , and s 3 to form an orthogonal space for user separation.
  • This step is called orthogonal construction.
  • a random symbol generator is used to generate the data to be sent in each round for transmission.
  • the receiving end receives the information, it first formats the data, and then user 1 uses its initial state value and 2 k cases of mi to perform sequential encoding, namely Execute the hash function with the initial state value to generate sa ,i , and then fill the initial state value into sa ,i for orthogonal construction to obtain s′ a,i (as mentioned above, the message M is divided into three symbols m 1 , m 2 and m 3 , each symbol has k bits, and the value of each bit is 0 and 1, so there are 2 k cases for each symbol).
  • s′ a,i is randomized to generate ra ,i by a random number generator
  • ra ,i is mapped to the constellation space to obtain x a,i .
  • the decoding process is basically the same as the encoding process at the beginning.
  • the receiving end performs sequential hash function, orthogonal construction, random number generation, and constellation mapping.
  • the generated data x a, i and The data received by User1 is compared and the Euclidean distance is calculated. After all mi calculations are completed, the data with the smallest Euclidean distance is selected as the decoding result.
  • VCG Vertical-Clarke-Groves
  • the principle of the VCG mechanism is that there is a set of participants and result sets, and each user evaluates the possible results.
  • the goal of the VCG mechanism is to select a set of results to maximize social benefits, and then each participant pays for the harm caused to other participants.
  • the VCG mechanism degenerates into the second highest price auction mechanism, that is, the highest bidder obtains the result he wants, but only pays the second highest price.
  • the VCG mechanism when used for spectrum auctions, multiple auctions are performed in each time slot, and only one carrier is auctioned each time, and the bid (the price can be an actual currency unit or a virtual
  • the user with the highest currency unit obtains the carrier to maximize his utility.
  • the utility of the user is the return of the unit price of each carrier (for example, network capacity or available rate, etc.).
  • each user creates an observation history for each carrier, and records the highest price of each auction in each time slot of the carrier. If a user is assigned the carrier, The highest price is the price paid by oneself; conversely, if a user has not been allocated the carrier, the highest price is the winner's price.
  • some learning algorithms such as Dirichlet Process (DP) and Gaussian process regression (GP regression), are used to obtain the predicted price of each carrier.
  • the user in order to maximize the user's own utility, the user allocates power based on the predicted price and channel gain. In other words, users allocate more power to users with relatively low prices and high channel gains; while users with too high prices and low channel gains do not allocate power.
  • a user can bid for multiple sub-carriers, but each sub-carrier can only be allocated to one user, and the power cannot exceed the maximum power of the user. Since the user’s bidding relies on subjective judgment (for example, the bid price is obtained after the user predicts the result of each auction as an observation value, it is subjective judgment.
  • the sub-carriers that the user finally allocates are only a part of all the sub-carriers bid, so after the auction ends, the user needs to reallocate the power and calculate their own return.
  • Figure 5 shows a spectrum allocation method based on auction theory according to an embodiment of the present invention. The method includes the following steps:
  • Step S510 the AP broadcasts carriers available for allocation.
  • the AP broadcasts the available carriers to all users for users to choose whether to bid and determine the bid price.
  • step S520 the user predicts the price based on historical observations.
  • Each user predicts the bid price based on the historical observation value, here the highest price of the historical auction of the observation value sub-carrier (see step S570 for details).
  • step S530 the user allocates power to each carrier according to the predicted price and channel quality to maximize its own return.
  • the user allocates power to each carrier based on the predicted price and channel quality (for example, statistical block error rate, bit error rate, etc.).
  • the pre-allocation of power is to decide whether to actually bid for a certain subcarrier. If the predicted price of the carrier is high and the channel quality is poor, no power is allocated, and if the channel quality is good and the predicted price is low, more power is allocated. In other words, the user decides whether to bid for the carrier based on the predicted price and channel quality, so as to maximize his own return, where the return refers to the capacity per unit price.
  • step S540 the user bids according to his needs.
  • users bid on the carrier that maximizes their return.
  • sub-carrier 1, sub-carrier 3, and sub-carrier 6 can maximize their returns, and then users will bid on these sub-carriers, and the specific bid amount can be determined by predicting the price.
  • step S550 the AP broadcasts the bid winner's price and allocates it to the winner carrier and the price to be paid.
  • the user sends the bid price to the AP, and the AP broadcasts the bid winner's price so that all users can understand the information of other users.
  • AP is allocated to the winner carrier and the price that needs to be paid.
  • step S560 the user allocates power again and calculates the reward according to the final bidding result.
  • step S570 the user records the highest price of each sub-carrier auction and updates the observed value. Since the AP will broadcast the bid winner price, each user can know the highest auction price of each subcarrier. Therefore, each user can create an observation history for each carrier and record the auction of each time slot of the carrier. The highest price is to update the observed value according to the auction result. If a user is assigned the carrier, the highest price is the price paid by oneself, and if a user is not assigned the carrier, the highest price is the winner's price.
  • users access the network through millimeter wave directional beams.
  • the beam width is adjusted first. , Determine the beam width when the throughput reaches the maximum and the users covered; then, the users access the network through the hash-based multiple access technology, in order to avoid collisions of data packets of different users, use constellation redundancy Construct an orthogonal space for user separation, so that users can access the AP at the same time in the uplink and downlink.
  • the sender encodes and sends the message according to the encoder, and the decoder at the receiver decodes the message correctly. In this way , Can combine with wavelength division multiplexing to increase network capacity.
  • the invention further multiplexes in a single beam on the basis of wavelength division multiplexing, so that users in a single beam can participate in concurrent transmission, and the spatial multiplexing rate is improved.
  • special encoders with randomness, orthogonality and linearity are used to adapt to the directional gain changes and communication link changes caused by the beamwidth adjustment.
  • the constellation diagram By fully accessing the constellation diagram, the 5G millimeter wave In the small cell network, an orthogonal space is constructed for robust multiple access.
  • the spectrum allocation strategy based on the bidding mechanism of the present invention can stimulate more users to participate in concurrent transmission and improve the transmission efficiency of users.
  • the present invention may be a system, a method and/or a computer program product.
  • the computer program product may include a computer-readable storage medium loaded with computer-readable program instructions for enabling a processor to implement various aspects of the present invention.
  • the computer-readable storage medium may be a tangible device that holds and stores instructions used by the instruction execution device.
  • the computer-readable storage medium may include, but is not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing, for example.
  • Computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM) Or flash memory), static random access memory (SRAM), portable compact disk read-only memory (CD-ROM), digital versatile disk (DVD), memory stick, floppy disk, mechanical encoding device, such as a printer with instructions stored thereon
  • RAM random access memory
  • ROM read-only memory
  • EPROM erasable programmable read-only memory
  • flash memory flash memory
  • SRAM static random access memory
  • CD-ROM compact disk read-only memory
  • DVD digital versatile disk
  • memory stick floppy disk
  • mechanical encoding device such as a printer with instructions stored thereon

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本发明提供一种毫米波通信方法和通信系统。该方法包括：对于毫米波传输，以吞吐量为优化目标确定优化的波束宽度和所述优化的波束宽度对应的用户；对于所述优化的波束宽度对应的多个用户，利用基于星座图冗余的多址接入进行并发传输。本发明提供的毫米波通信方法和通信系统通过对波束宽度进行自适应调节和改进多址接入技术能够提高网络性能，改善用户体验。

Description

一种毫米波通信方法和通信系统 技术领域
本发明涉及无线通信技术领域，尤其涉及一种毫米波通信方法和通信系统。
背景技术
随着移动通信需求的爆炸性增长，容量需求和频谱短缺之间的矛盾日益突出。无线带宽瓶颈成了无线网络的关键问题。在这种情况下，毫米波以其独特的带宽优势和非授权频谱可以很好地解决“带宽饥饿”问题，并为通信网络提供吉比特级的通信需求以及超低延迟传输。
然而，毫米波网络并不是低频网络的简单升级。根据Friis传输方程(弗里斯传输方程)，自由空间传播路径的损耗与频率的二次方成正比。与低频段相比，毫米波面临巨大的传输损耗。为了解决这一问题，毫米波采用波束成形技术，产生定向增益来弥补这种传输损耗，但是这种定向窄波束传输存在的问题有：(1)采用定向传输的毫米波收发机需要在建立有效链路之前进行相应的波束扫描和对准，但此过程的时间复杂度和空间复杂度较高，一旦链路发生了变化(例如接收机的位置改变)，会使得原来的通信链路不可用，只能进行新一轮的波束扫描过程，时间开销巨大。(2)这种定向传输的波束宽度通常极窄，一旦遇到障碍物，例如人的走动，会使整个通信链路中断。(3)在两个终端进行通信时，第三方节点不能够进行载波监听，从而使第三方节点的退避时间过长。(4)由于毫米波阵列天线单元数量有限，所以产生的通信波束也有限，所以不能够很好的利用毫米波空间资源进行并发传输。
受限于毫米波频段的传输特性，毫米波通信的研究与应用主要集中在点对点链路，为了进一步开发毫米波的频谱资源，利用空间自由度是另一种满足日益增长的数据速率和服务质量的方法。例如，通过在发射端和接收端部署多根天线，形成多输入多输出(MIMO)系统，这种传输方法实现了空间分集，复用增益以及容量提升。在多用户多输入多输出系统 (MU-MIMO)或大规模多输入多输出系统(massive MIMO)中，通过部署大规模阵列天线产生多个窄波束能够提高波束复用增益并减小用户间干扰，使得接入设备(例如基站、接入点等)能够进一步利用空间自由度服务多个移动用户。但大规模多输入多输出系统的高吞吐量依赖于基站端的信道状态信息，而在频分复用系统中，由于大规模的天线使得信道状态信息的反馈开销大，在这种情况下，有人提出使用波分复用(Beam Division Multiple Access，BDMA)传输模型，使得基站通过不同的波束与用户通信，同时将用户分离到不重叠的波束中，因而多个用户可以同时接入信道。
综上，在现有技术中，尽管毫米波通信网络能够采用波束成形、多输入多输出(MIMO)和波分复用等技术来提高网络性能，但存在不能自适应调节波束宽度、不能自适应通信链路的变化等问题，从而影响了资源利用率和用户体验。
因此，需要对现有技术进行改进，以提供适用于毫米波通信的通信方法。
发明内容
本发明的目的在于克服上述现有技术的缺陷，提供一种毫米波通信方法和通信系统。
根据本发明的第一方面，提供一种毫米波通信方法。该方法包括以下步骤：
步骤S1：对于毫米波传输，以吞吐量为优化目标确定优化的波束宽度和所述优化的波束宽度对应的用户；
步骤S2：对于所述优化的波束宽度对应的多个用户，利用基于星座图冗余的多址接入进行并发传输。
在一个实施例中，步骤S1包括以下子步骤：
设置最小波束宽度、最大波束宽度以及波束宽度变化量；
以所述波束宽度变化量为步长，在所述最小波束宽度和所述最大波束宽度之间进行扫描，找出系统最大吞吐量时对应的波束宽度，作为所述优化的波束宽度。
在一个实施例中，在步骤S2包括以下子步骤：
基于哈希运算在消息比特和编码比特之间产生随机、非线性的映射；
将码比特映射到星座符号集以获得正交的编码。
在一个实施例中，对于一个用户，所述哈希运算包括：
将该用户消息分成多个数据块，对每个数据块分别执行哈希运算，并且以前一个数据块执行哈希运算生成的结果作为后一个数据块执行哈希运算的输入。
在一个实施例中，在步骤S2中，对于所述优化的波束宽度对应的多个用户，基于拍卖理论进行载波分配。
在一个实施例中，步骤S2包括以下子步骤：
用户接收可分配载波并根据竞拍的历史观察值对所述可分配载波进行价格预测，获得可分配载波的预测价格；
用户根据所述可分配载波的预测价格和信道质量为所述可分配载波分配功率；
用户确定将进行竞拍的载波和投标价格；
用户接收竞标结果，其中，所述竞拍结果包括赢家价格。
在一个实施例中，对可分配载波进行价格预测包括：
每个用户记录下每个载波的拍卖历史最高价作为观察值，并根据观察值对下一时隙子载波价格进行预测，决定投标价格。
根据本发明的第二方面，提供了一种毫米波通信系统。该系统包括：
接入设备：对于毫米波传输，以吞吐量为优化目标确定优化的波束宽度和所述优化的波束宽度对应的用户。
用户终端：对于所述优化的波束宽度对应的多个用户，利用基于星座图冗余的多址接入进行并发传输。
在一个实施例中，所述接入设备用于执行：设置最小波束宽度、最大波束宽度以及波束宽度变化量；以所述波束宽度变化量为步长，在所述最小波束宽度和所述最大波束宽度之间进行扫描，找出系统最大吞吐量时对应的波束宽度，作为所述优化的波束宽度。
与现有技术相比，本发明的优点在于：对于毫米波传输，利用可变的波束宽度激励更多的并发传输，利用无速率码来适应信道变化，采用新的多址接入技术进行并发传输。此外，利用拍卖理论策略性地分配频谱资源和用户自身的功率，从而提高用户的传输效益，解决了毫米波通信高度定向性传输导致地空间复用能力低的问题。
附图说明
以下附图仅对本发明作示意性的说明和解释，并不用于限定本发明的范围，其中：
图1示出了根据本发明一个实施例的毫米波通信系统的模型图；
图2示出了根据本发明一个实施例的波束宽度自适应调节方法的流程图；
图3示出了根据本发明一个实施例的基于星座图冗余的多址接入体系结构；
图4示出了根据本发明一个实施例的编解码过程示意图；
图5示出了根据本发明一个实施例的基于拍卖理论的频谱分配方法的流程图。
具体实施方式
为了使本发明的目的、技术方案、设计方法及优点更加清楚明了，以下结合附图通过具体实施例对本发明进一步详细说明。应当理解，此处所描述的具体实施例仅用于解释本发明，并不用于限定本发明。
在本文示出和讨论的所有例子中，任何具体值应被解释为仅仅是示例性的，而不是作为限制。因此，示例性实施例的其它例子可以具有不同的值。
对于相关领域普通技术人员已知的技术、方法和设备可能不作详细讨论，但在适当情况下，所述技术、方法和设备应当被视为说明书的一部分。
本发明提供的实施例在毫米波通信网络中，既能够允许多个用户接入，又能够根据信道质量保证可靠性传输，从而提高空间复用效率。此外，为进一步最大化用户的传输效益，本发明实施例还考虑频谱分配和功率分配问题。以下将重点介绍波束宽度调节、多址接入编解码、频谱分配和功率分配的实施例。
(1)波束宽度调节的实施例
为了允许多个用户接入，本发明实施例对波束宽度进行自适应调节。例如，参见图1所示，对于一个AP(接入点)和多个用户的情况，用户随机分布在AP附近，其中，AP通过阵列天线单元产生多个定向波束来服务用户。在密集的用户场景中，一个波束可能覆盖多个用户。假设在波束内AP可以同时和多个用户进行通信，通过理想扇形天线模型可以知道波束的方向增益和波束宽度的关系，即波束宽度越宽，波束的方向增益越小； 相反，波束的宽度越窄，波束的方向增益越大。波束的方向增益可以提高用户的信噪比，根据香农公式，信噪比越大，传输速率越大，因此，波束方向增益越大，用户的传输速率也越大，波束宽度越窄，而波束宽度越窄，覆盖的用户数越少。从整体来看，加大波束宽度可以提高用户覆盖数量，从而提高系统吞吐量，但单个用户传输速率不能保证。相比之下，减小波束宽度则可以提高单个用户的传输速率，从而提高系统吞吐量，但用户覆盖数量可能降低。基于上述原理，存在一个最大吞吐量，而实现最大吞吐量时的波束宽度，在本发明中称为最优的波束宽度。
在一个实施例中，以最大系统吞吐量为优化目标，寻找最优的波束宽度。例如，从最小的波束宽度开始依次增加波束宽度，每次增加波束宽度后计算一次系统吞吐量，再与前一次计算的系统吞吐量进行比较，将获得最大系统吞吐量时对应的波束宽度作为最优的波束宽度。
具体地，参见图2所示，本发明实施例提供的波束宽度自适应调节方法包括以下步骤：
步骤S210，设置最小波束宽度并计算对应的吞吐量。
在此步骤中，可根据天线阵列的规模、用户对传输质量的需求、用户密集程度等设置最小波束宽度和最大波束宽度。进一步地，计算最小波束宽度时对应的系统吞吐量。
步骤S220，增加波束宽度并计算对应的吞吐量。
在此步骤中，每次将波束宽度增加一定数量，并计算对应的系统吞吐量，其中，波束宽度每次增加的数量可根据执行速度和波束宽度分辨率设置，例如，如果需要快速确定最优波束宽度，则每次增加较大数量的波束宽度，而如果需要确定较精确的最优波束宽度，则每次增加较少数量的波束宽度。
步骤S230，比较是否大于最大波束宽度。
在此步骤中，将当前波束宽度对应的系统吞吐量和前一次波束宽度对应的系统吞吐量进行比较，如果大于，并且当前波束宽度已达到设置的最大波束宽度，则执行步骤S240，如果当前波束宽度未达到设置的最大波束宽度，则继续执行步骤S220。
步骤S240，确定吞吐量最大时的波束宽度和用户。
当在设置的最小波束宽度和最大波束宽度范围内扫描完成之后，找出系统最大吞吐量对应的波束宽度，记录为最优波束宽度并记录此时覆盖的 用户，可以是一个用户或多个用户。
需要说明的是，本领域的技术人员也可从设置的最大波束宽度进行扫描，不断降低波束宽度，直到找出系统最大吞吐量对应的波束宽度。
(2)多址接入编解码器的实施例。
本发明实施例提供了多个用户进行并发传输的多址接入技术，这种多址接入技术能够自适应信道状态变化并达到以下目的：在下行链路，AP和一组用户同时进行通信，AP将所有消息编码成一组无速率码，选择一个比链路条件允许的最大速率更高的传输速率广播出去，一旦目的用户接收到这组无速率码，各用户持续从这组码中提取出有用的信息，直到自身的消息被成功解码为止，然后每个用户发送一个反馈消息给AP，AP接收到所有反馈消息后再进行下一轮传输。在上行链路，各用户将自己的数据包编码成一组无速率码，在每次传输时，这些用户同时将无速率码以一个固定速率发送给AP，一旦AP接收到这组无速率码，按照用户将消息分离解码出来，一旦成功解码，AP再反馈一个反馈消息给相应的用户，用户接收到反馈消息后，将进行下一轮传输。为了达到这个目的，实现用户的多址接入，本发明提供了改进的编码器和解码器。
在一个实施例中，采用基于哈希的编码器。具体地，该编码器包括顺序编码器、随机数生成器(也称为随机符号生成器)、正交构造器，其中，顺序编码器和随机数生成器用于维持鲁棒性的传输链路，正交构造器用于提供星座空间内的并发传输。这种编码器具有随机性、正交性、线性等特性，随机性是利用密集星座图传输的核心，对噪声和干扰的弹性将显著提高；正交性是保证多用户分离的结构，在星座图空间中构建正交结构是十分必要的；线性是能够实现具有多项式时间复杂度和空间复杂度解码器的关键，同时对于多用户场景下的线性解码器也是不可或缺的。
星座图通常代表向量空间中的调制信号，每个信号在星座图中都有一个位置，叫做星座点。不同的调制模式指示所有可能的星座点的位置，不同的调制模式使星座点之间的距离不同。星座图越稠密调制阶数越高，能够实现的数据传输率也越高，星座点之间的距离也越近，这时需要更好的信道条件才能解码出正确的信号。无线网络通常采用固定且对称的星座图，例如卷积码和低密度奇偶校验码，但在信道质量差的情况下，星座点的固定距离会使速率传输率受限。本发明实施例采用哈希函数的多址接入技术继承了无速率码的几个优点。
具体而言，编码器采用了哈希函数，在消息比特和编码比特之间产生随机、非线性的映射，然后将码位(码比特)映射到密集星座符号集中来尽可能传送更多的信息。因为哈希函数本身的特性，使得旧的状态码和新的状态码长度一致，而且对于任何两个输入信息，即使仅有一个比特不同，生成的结果也完全不同，通过这种方式，可以有效对抗噪声和干扰。
进一步地，为了使更多用户进行传输，本发明实施例的毫米波多址接入技术构建了一个正交编码比特用于用户分离。首先将一组初始状态值(发送端的编码器和接收端的解码器互相知道，例如所有的值全部为0)分为几个段，每个段分配给一个用户。将用户的消息分成几个小块，然后对消息块执行顺序哈希函数，每次执行哈希函数之后将会产生一个结果，由于哈希函数的性质，使得这些结果可以构成一个正交空间。当接收到初始状态值之后，将初始状态值零填充到正交空间中，补充每个消息块的长度。通过这种方式，来自并发传输的用户在星座维度是正交的，从而能够正确解码。
在正交化之后，进一步随机化以维持无速率码的性质。然而，在现有技术的无速率编码结构中，正交性被非线性随机发生器(Random Number Generator，RNG)破坏。此外，任何两个码字的聚合不总是一个逻辑码字，因为编码是非线性的。对于多用户同时传输的场景下的解码算法而言，计算是不切实际的。为了解决这个问题，本发明实施例使用线性同余生成器(Linear congruential Generator，LCG)或其变体来替换非线性随机发生器(RNG)进行随机化，这确保了多用户场景中解码器的实用性。在适当的参数下，例如模数和乘数，线性同余生成器(LCG)的通用随机性和独立性与非线性随机发生器(RNG)相当。因此，无速率特性不再与正交性和线性要求相矛盾。
相应地，本发明实施例的毫米波多址接入的解码算法从编码结构中获得了优势，例如，通过搜索一棵基于最大似然(Maximum Likelihood，ML)的树，以找到最小花费的分支为目标。然而，计算复杂度随着并发传输的数量呈指数增长，关键的挑战是如何实现具有确保性能和最小计算开销的ML解码器。由于编码器的线性特性，接收端只需要构建自身的解码树，解码结果是最小花费的分支。为了提高解码速度，在一个实施例中，采用基于软信息的M-算法来修剪树，从而进一步降低了解码复杂度。众所周知，最大似然解码器有着指数级的解码复杂度，随着多址接入数量的增加， 通过简单地使用M-算法来剪枝是不可能实现的，因为有时可能会由于失真将正确的码丢弃，但是丢弃的码中携带着大量可靠的软信息，通过提取这些软信息可以提高解码精确度以及减小计算开销。这些软信息叫做“对数似然比”(Log-likelihood Ratio，LLR)，指示了解码器对每个比特决策的确定性，用于纠正最终解码决策。
理论上，一个N阶调制的星座图可以支持的并发传输的数量为N，这是因为星座图上潜在N个正交星座点。随着调制阶数增高，正确解码就需要更好的信道条件，所以，最大可以支持的并发传输的数量不可能超过香农极限。随着并发传输的增加，每个用户的吞吐量就会降低，所以实际上，需要在并发传输和每个用户的吞吐量上找到平衡。
图3示出了根据本发明一个实施例的基于毫米波的通信系统，该系统包括发送端和接收端，发送端的处理包括加扰、编码器、正交构造器、随机符号生成器、星座图映射器，经过星座图映射之后，生成的调制信号经过物理层的其他处理并经由RF前端发送到接收端，与发送端对应，接收端的处理包括星座图解映射、多用户检测、顺序解码器、解扰等。
需要说明的是，图3示意的是通信网络的数据发送、接收的典型过程，该过程属于现有技术，例如，加扰、解扰、CRC校验、发送过程的重试计数等，在此不再赘述。本发明的贡献在于，提供的编码器包括顺序编码器和随机符号生成器来维持传输链路的鲁棒性，正交构造器则在星座空间提供并发传输；星座解码器基于软信息和最大似然估计实现具有最小解码复杂度的实际解码器。此外，本发明通过波束宽度自适应调节来适应链路质量和用户需求。
具体地，本发明实施例提供的编解码过程参见图4所示，以user1(用户1)为例，将User1的消息M分为三个不同的符号m 1、m 2、m 3，每个符号有k比特，用户1的初始状态与第一个符号m 1经过哈希编码器(标记为h)生成s 1，s 1和m 2经过哈希编码器生成s 2，这一步称为循序编码。随后，将初始状态值填充至s 1、s 2和s 3中产生s′ 1、s′ 2、s′ 3，组成一个正交空间，用于用户分离，这一步称为正交构造。最后，利用随机符号生成器生成每一轮要发送的数据进行发送。接收端收到信息后，首先，将数据进行格式化，然后用户1利用自己的初始状态值和2 k种情况的m i进行顺序编码，即
Figure PCTCN2019081914-appb-000001
和初始状态值执行哈希函数后生成s a,i，再将初始状态值填充至s a,i中进行正交构造，得到s′ a,i(上文提到，消息M分成三个符号m 1、m 2和m 3， 每个符号有k个比特，其中每个比特的取值为0和1，所以每个符号就有2 k种情况)。接下来，将s′ a,i通过随机数生成器随机化生成r a,i，最后将r a,i映射到星座空间中得到x a,i。由图4可以看出解码过程一开始和编码过程基本一致，接收端根据初始状态码，执行顺序哈希函数、正交构造、随机数生成以及星座映射，最后将生成的数据x a,i与User1接收到的数据进行比较，计算欧氏距离，在所有的m i计算完毕之后，选取最小欧式距离的数据作为解码结果。
(3)频谱分配和功率分配的实施例
在经过上述的物理层编解码之后，用户可以根据多址接入技术进行并发传输。为了进一步优化并发传输用户的传输效益，在一个实施例中，采用基于VCG(Vickrey–Clarke–Groves)拍卖的频谱分配机制。VCG机制的原理是：有一组参与者和结果集，每个用户为可能的结果进行估价。VCG机制的目标是选择一组结果使得社会效益最大化，之后每个参与者为对其他参与者造成的伤害付款。当结果集只有一个时，VCG机制就退化为第二高价拍卖机制，即出价最高者获得自己想要的结果，但只需支付第二高价。
在本发明的一个实施例中，当采用VCG机制进行频谱拍卖时，每个时隙进行多次拍卖，每次只拍卖一个载波，出价(价格可以是实际上的货币单位，也可以是一个虚拟货币单位)最高的用户获得该载波，使得自己的效用最大，用户的效用为每个载波的单位价格的回报(例如网络容量或者可以获得的速率等)。进一步地，为了使用户更好的决策投标价格，每个用户为每个载波创建一个观察历史，记录下该载波每个时隙中每次拍卖的最高价，如果一个用户被分配了该载波，则最高价为自己的支付价格；反之，如果一个用户未被分配该载波，则最高价为赢家价格。根据这些观察值，使用一些学习算法，例如狄利克雷过程(Dirichlet Process，DP)和高斯过程回归(Gaussian process regression，GP regression)，获得每个载波的预测价格。
在一个实施例中，为了最大化用户自身的效用，用户根据预测价格和信道增益来分配功率。换句话说，用户为价格相对低而信道增益高的用户分配更多功率；而价格过高且信道增益过低的不分配功率。需要说明的是，一个用户可以竞拍多个子载波，但每个子载波只能分配给一个用户，且功率不能超过用户的最大功率。由于用户的出价依赖于主观判断(例如所出的价是用户将每次拍卖结果作为观察值进行预测后得出的，所以是主观判 断的，实际上用户根据预测价格是否能获得拍卖胜利是不确定的。)，用户最终分配到的子载波只是投标的所有子载波中的一部分，所以拍卖结束后，用户需要重新分配功率并计算自身的回报。
图5示出了本发明一个实施例的基于拍卖理论的频谱分配方法，该方法包括以下步骤：
步骤S510，AP广播可供分配的载波。
在此步骤中，AP将可供分配的载波广播给所有用户，以供用户选择是否进行投标并确定投标价格。
步骤S520，用户根据历史观察值预测价格。
每个用户根据历史观察值对投标价格进行预测，这里的观察值子载波的历史竞拍的最高价(具体参见步骤S570)。
步骤S530，用户根据预测价格和信道质量为各载波分配功率，使自身回报最大。
在此步骤中，用户根据预测价格和信道质量(例如，统计的误块率、误码率等)给每个载波分配功率，预先分配功率是为了决定实际上是否对某个子载波进行投标。如果该载波预测价格高且信道质量差，则不分配功率，如果信道质量好并且预测价格较低，则分配较多功率。换句话说，用户根据预测价格和信道质量来决定是否给该载波进行投标，使得自身的回报最大，这里的回报指单位价格的容量。
步骤S540，用户根据自身需要出价竞标。
在一个实施例中，用户对使自己回报最大的载波进行出价竞标。例如，子载波1、子载波3以及子载波6可以使得自身的回报最大，那么用户将对这些子载波出价竞标，具体出价多少可通过预测价格确定。
步骤S550，AP广播竞拍的赢家价格并分配给赢家载波和需要支付的价格。
用户将投标价格发送给AP，AP广播竞拍的赢家价格，以使所有用户了解其他用户的信息。此外，AP分配给赢家载波和需要支付的价格。
步骤S560，用户根据最终的投标结果再次分配功率和计算回报。
在AP为赢家分配载波和需要支付的价格的过程中，由于一些用户对所竞拍的载波竞拍失败，所以用户需要根据实际竞拍的载波重新分配功率和计算回报。。
步骤S570，用户记录下各子载波每次拍卖的最高价并更新观察值。由 于AP会广播竞拍的赢家价格，每个用户都可以获知各个子载波的拍卖最高价，因此，各用户可以为每个载波创建一个观察历史，记录下该载波每个时隙中每次拍卖的最高价，即根据竞拍结果更新观察值，如果一个用户被分配了该载波，则最高价为自己的支付价格，而如果一个用户未被分配该载波，则最高价为赢家价格。
综上所述，在本发明实施例中，用户通过毫米波定向波束接入网络，为了使在单个波束内可以同时进行多个用户并发传输，以提高波束内的吞吐量，首先进行波束宽度调节，确定吞吐量达到最大时的波束宽度和所覆盖的用户；然后，用户通过基于哈希的多址接入技术接入到网络中，为避免不同用户的数据包发生冲突，利用星座图冗余构造一个正交空间用于用户分离，以使用户在上行链路和下行链路同时接入AP，发送端将消息根据编码器进行编码发送，接收端的解码器将消息正确解码，通过这种方式，能够结合波分复用提高网络容量。
本发明通过波束宽度自适应调节，在波分复用的基础上，在单个波束内进一步复用，使单个波束内的用户可以参与到并发传输，提高了空间复用率。在波束宽度自适应调节之后，通过具有随机性、正交性和线性属性的特殊编码器来适应波束宽度调节引起的方向增益变化和通信链路变化，通过充分接入星座图，在5G毫米波小蜂窝网络中为鲁棒性的多址接入构建正交空间。进一步地，本发明基于竞拍机制的频谱分配策略可以激发更多的用户参与到并发传输并提高了用户的传输效益。
需要说明的是，虽然上文按照特定顺序描述了各个步骤，但是并不意味着必须按照上述特定顺序来执行各个步骤，实际上，这些步骤中的一些可以并发执行，甚至改变顺序，只要能够实现所需要的功能即可。
本发明可以是系统、方法和/或计算机程序产品。计算机程序产品可以包括计算机可读存储介质，其上载有用于使处理器实现本发明的各个方面的计算机可读程序指令。
计算机可读存储介质可以是保持和存储由指令执行设备使用的指令的有形设备。计算机可读存储介质例如可以包括但不限于电存储设备、磁存储设备、光存储设备、电磁存储设备、半导体存储设备或者上述的任意合适的组合。计算机可读存储介质的更具体的例子(非穷举的列表)包括：便携式计算机盘、硬盘、随机存取存储器(RAM)、只读存储器(ROM)、可擦式可编程只读存储器(EPROM或闪存)、静态随机存取存储器(SRAM)、 便携式压缩盘只读存储器(CD-ROM)、数字多功能盘(DVD)、记忆棒、软盘、机械编码设备、例如其上存储有指令的打孔卡或凹槽内凸起结构、以及上述的任意合适的组合。
以上已经描述了本发明的各实施例，上述说明是示例性的，并非穷尽性的，并且也不限于所披露的各实施例。在不偏离所说明的各实施例的范围和精神的情况下，对于本技术领域的普通技术人员来说许多修改和变更都是显而易见的。本文中所用术语的选择，旨在最好地解释各实施例的原理、实际应用或对市场中的技术改进，或者使本技术领域的其它普通技术人员能理解本文披露的各实施例。

Claims (10)

  1. 一种毫米波通信方法，包括以下步骤：
    步骤S1：对于毫米波传输，以吞吐量为优化目标确定优化的波束宽度和所述优化的波束宽度对应的用户；
    步骤S2：对于所述优化的波束宽度对应的多个用户，利用基于星座图冗余的多址接入进行并发传输。
  2. 根据权利要求1所述的方法，其中，步骤S1包括以下子步骤：
    设置最小波束宽度、最大波束宽度以及波束宽度变化量；
    以所述波束宽度变化量为步长，在所述最小波束宽度和所述最大波束宽度之间进行扫描，找出系统最大吞吐量时对应的波束宽度，作为所述优化的波束宽度。
  3. 根据权利要求1所述的方法，其中，在步骤S2包括以下子步骤：
    基于哈希运算在消息比特和编码比特之间产生随机、非线性的映射；
    将码比特映射到星座符号集以获得正交的编码。
  4. 根据权利要求3所述的方法，其中，对于一个用户，所述哈希运算包括：
    将该用户消息分成多个数据块，对每个数据块分别执行哈希运算，并且以前一个数据块执行哈希运算生成的结果作为后一个数据块执行哈希运算的输入。
  5. 根据权利要求1所述的方法，其中，在步骤S2中，对于所述优化的波束宽度对应的多个用户，基于拍卖理论进行载波分配。
  6. 根据权利要求5所述的方法，步骤S2包括以下子步骤：
    用户接收可分配载波并根据竞拍的历史观察值对所述可分配载波进行价格预测，获得可分配载波的预测价格；
    用户根据所述可分配载波的预测价格和信道质量为所述可分配载波分配功率；
    用户确定将进行竞拍的载波和投标价格；
    用户接收竞标结果，其中，所述竞拍结果包括赢家价格。
  7. 根据权利要求6所述的方法，其中，对可分配载波进行价格预测包括：
    每个用户记录下每个载波的拍卖历史最高价作为观察值，并根据观察 值对下一时隙的子载波价格进行预测，决定投标价格。
  8. 一种毫米波通信系统，包括：
    接入设备：对于毫米波传输，以吞吐量为优化目标确定优化的波束宽度和所述优化的波束宽度对应的用户。
    用户终端：对于所述优化的波束宽度对应的多个用户，利用基于星座图冗余的多址接入进行并发传输。
  9. 根据权利要求8所述的系统，其中，所述接入设备用于执行以下过程：
    设置最小波束宽度、最大波束宽度以及波束宽度变化量；
    以所述波束宽度变化量为步长，在所述最小波束宽度和所述最大波束宽度之间进行扫描，找出系统最大吞吐量时对应的波束宽度，作为所述优化的波束宽度。
  10. 一种计算机可读存储介质，其上存储有计算机程序，其中，该程序被处理器执行时实现根据权利要求1至7中任一项所述方法的步骤。
PCT/CN2019/081914 2019-04-09 2019-04-09 一种毫米波通信方法和通信系统 WO2020206604A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2019/081914 WO2020206604A1 (zh) 2019-04-09 2019-04-09 一种毫米波通信方法和通信系统

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2019/081914 WO2020206604A1 (zh) 2019-04-09 2019-04-09 一种毫米波通信方法和通信系统

Publications (1)

Publication Number Publication Date
WO2020206604A1 true WO2020206604A1 (zh) 2020-10-15

Family

ID=72751754

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/081914 WO2020206604A1 (zh) 2019-04-09 2019-04-09 一种毫米波通信方法和通信系统

Country Status (1)

Country Link
WO (1) WO2020206604A1 (zh)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107453795A (zh) * 2017-08-21 2017-12-08 东南大学 多用户毫米波通信系统的波束分配方法及其装置和系统
CN109547080A (zh) * 2018-11-30 2019-03-29 华东师范大学 一种基于stbc有效补偿波束偏斜的模拟波束传输方法

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107453795A (zh) * 2017-08-21 2017-12-08 东南大学 多用户毫米波通信系统的波束分配方法及其装置和系统
CN109547080A (zh) * 2018-11-30 2019-03-29 华东师范大学 一种基于stbc有效补偿波束偏斜的模拟波束传输方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
WANG, LU ET AL.: "Multiple Access MmWave Design for UAV-Aided 5G Communications", IEEE WIRELESS COMMUNICATIONS, vol. 26, no. 1, 13 February 2019 (2019-02-13), XP011709554, DOI: 20191210152634X *

Similar Documents

Publication Publication Date Title
CN110048755B (zh) 一种毫米波通信方法和通信系统
US10924212B2 (en) System and method for user equipment cooperation
EP3245745B1 (en) System and method for a message passing algorithm
Shirvanimoghaddam et al. Massive multiple access based on superposition raptor codes for cellular M2M communications
US8824582B2 (en) Base station and method for channel coding and link adaptation
US12009839B2 (en) Capacity achieving multicarrier modulation and coding systems and methods
RU2733826C1 (ru) Высокоскоростные длинные ldpc коды
WO2018201481A1 (en) Method and device for incremental redundancy hybrid automatic repeat request (ir-harq) re-transmission
JP2009303197A (ja) 低密度パリティコードエンコーディング／デコーディング装置及びその方法並びにコンピュータ読み取り可能な記録媒体
WO2020147526A1 (zh) 一种级联crc码的极化码编码方法及装置
WO2019238117A1 (zh) 重复传输方法和通信装置
KR102482876B1 (ko) Mimo 채널에 대한 폴라 코드 생성 장치 및 방법
RU2671954C1 (ru) Способ, оборудование и устройство для определения порядка модуляции и кодирования
US11451244B2 (en) Device and method for encoding and decoding using polar code in wireless communication system
CN112994869B (zh) 一种通信系统、方法、设备及存储介质
WO2021136400A1 (zh) 上行信号压缩传输方法、系统、计算机设备和存储介质
Mahmoodi et al. Non-symmetric multi-antenna coded caching for location-dependent content delivery
WO2020206604A1 (zh) 一种毫米波通信方法和通信系统
CN110611525A (zh) 一种基于速率分拆的信号传输、接收方法及装置
WO2018141271A1 (zh) 数据处理的方法和装置
Nasraoui et al. Throughput maximization with optimum energy allocation for ARQ retransmission protocol
US11316613B2 (en) Method of transceiving signal by using polar code and device for performing the method
Allu et al. Towards Improved Spectral Efficiency Using RSMA-Integrated Full-Duplex Communications
Mheich et al. Achievable rates optimization for broadcast channels using finite size constellations under transmission constraints
WO2022117061A1 (zh) 一种极化码辅助比特的确定方法和装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19924313

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205 DATED 25.01.2022)

122 Ep: pct application non-entry in european phase

Ref document number: 19924313

Country of ref document: EP

Kind code of ref document: A1