CN100370766C - Design Method of Test Platform for Distributed Wireless Communication Transmission Technology - Google Patents

Abstract

The present invention relates to a design method for a test platform of distributed wireless communication transmission technology, which can evaluate general wireless communication transmission technology and the important transmission technologies of multiple-input multiple-output communication systems in future wireless communication. Sockets are used for connecting processors of a test platform through the TCP/IP network protocol, and operations of an entire test are dispersed to two computers. On one hand, simulation tests are improved in the two aspects of simulation scale and execution speed; on the other hand, the physical separation of a transmitter from a receiver is really realized, and data received by the receiver is completely decided by a third party evaluation unit; as a result, the situation that a test party causes the short circuit of the transmitter and the receiver is completely stopped. Therefore, the present invention can increase the efficiency of the test and the evaluation of the wireless communication transmission technology, and the test fairness is ensured.

Description

Design Method of Test Platform for Distributed Wireless Communication Transmission Technology

technical field

The invention relates to a design method for a wireless communication transmission technology test platform, in particular to a design method for a distributed wireless communication transmission technology test platform based on Ethernet.

Background technique

In the past two decades, the wireless communication system represented by mobile communication has developed rapidly, and the first and second generation mobile communication systems represented by low-speed voice services have been increasingly unable to meet people's needs for data and multimedia services. The third-generation mobile communication system with high-speed data, multimedia, and intelligent services as the main goal is getting closer to people's lives. The third-generation mobile communication system is an important development stage for the evolution of future personal communication, and has epoch-making importance. It is bound to have a significant impact on human life and economic development. At the same time, with the gradual commercialization of the third-generation mobile communication system, the research on Beyond 3G and fourth-generation mobile communication at home and abroad has flourished.

Since the research and development of the second-generation mobile communication system, computer simulation testing has become the most powerful weapon for mobile communication research and development. Before each new technology or each new system is put into commercial use, many companies or research institutions conduct long-term simulations to test their performance. For the pre-research technical evaluation of the fourth-generation wireless communication system, it is necessary to use computer simulation methods to obtain objective verification before entering the actual measurement stage. Common simulation tests of communication systems can be divided into link-level simulation tests and system-level simulation tests. The system-level simulation test is mainly for the simulation test of the wireless access system. The link-level simulation test is mainly to simulate the wireless transmission technology (such as coding, interleaving, modulation and spread spectrum, etc.) of the physical layer of the communication system to obtain the performance of the technology under different wireless channels. The simulation results are mainly shown as the relationship curves of bit error rate (BER) or frame error rate (FER) and signal-to-noise ratio under different wireless environments. Link-level simulation has two main purposes, one is to test the performance of wireless transmission technology, and the other is to test the end-to-end performance of the wireless system and provide data support for system-level simulation.

Multiple-Input Multiple-Output (MIMO, Multiple-Input Multiple-Output) wireless communication transmission technology has good anti-fading and anti-noise capabilities, so that huge capacity can be obtained. Therefore, in a wireless channel with limited power bandwidth, multiple-input multiple-output (MIMO, Multiple-Input Multiple-Output) wireless communication transmission technology is one of the important technologies to achieve high data rate, improve system capacity and transmission quality. Its superior characteristics make it one of the key technologies in the future super three-generation communication system (B3G) and the fourth-generation communication system.

However, the amount of computing required for the evaluation of the wireless transmission technology of the MIMO communication system is much larger than that of the evaluation test of the previous SIMO communication system. If the test under multi-user conditions is considered, the amount of calculation cannot be realized by using an existing single high-performance computer.

Contents of the invention

The purpose of the present invention is to overcome the above-mentioned shortcoming of prior art, provide a kind of efficiency that can improve the test evaluation of wireless communication transmission technology, guarantee the design method of the distributed wireless communication transmission technology test platform of test fairness,

In order to achieve the above object, the technical scheme adopted in the present invention is:

1) Generation of source bit stream

First, use iterative centering method, shift method, multiplication congruence method or mixed congruence method to transform a random input binary sequence to generate a random signal sequence with complex Gaussian distribution, and then pass the random signal hypothesis testing method Generate a random source bit stream with a rate and frame length that is uniformly distributed on the (0, 1) interval and can be freely adjusted between 8Kbps and 100Mbps;

2) Link the transmitter module under test

According to the interface specification, the transmitter module under test is linked to the transmitter host of the test platform, and then the generated source bit stream is connected to the transmitter module under test, and the transmitter module under test performs channel coding, interleaving, modulation, Data output by space-time coding and framing algorithms;

3) Generation of multiple-input multiple-output MIMO channel modules

Assuming that a multiple-input multiple-output MIMO channel has M transmit antennas and N receive antennas, that is, M inputs and N outputs, then the MIMO channel is equivalent to M×N independent single-input single-output SISO channels, each independent The single-input single-output SISO channel has two characteristics of frequency selective fading and time selective fading. The realization of the frequency selective fading characteristic is to change a "broadband" signal into a "narrowband" signal through a multipath channel, and then pass through multiple Independent single-path channel weighting to obtain the required frequency selective fading, and the realization of time selective fading characteristics is to use the method of shaping filter to generate Doppler power spectrum shape, the amplitude obeys the Rayleigh distribution, and the phase obeys the uniform distribution The random process sampling signal is multiplied by the output signal of the transmitter under test to obtain the required time-selective fading, that is, M×N single-input single-output SISO channels are generated, and then multiple-input multiple-output MIMO channel;

4) Verification of the independence of multiple-input multiple-output MIMO channel modules

Assuming that the number of multipath channels between any one of the M×N single-input single-output SISO channels is K, randomly sample M×N×K independent single-path channels and select any two groups of them to form the corresponding Two sets of random variables, use the correlation function or cross-covariance function to test the independence between the two channels, if the value of the two correlation functions is zero, the two channels are independent of each other, otherwise transfer to the MIMO channel Module generation regenerates the multiple-input multiple-output MIMO channel module;

5) Network transmission of data

After the data output by the transmitting host passes through the MIMO channel, the socket is used to realize the TCP/IP protocol network transmission, that is, the data is transmitted to the receiving host, and the receiver module under test in the receiving host reads the data;

6) Link the receiver module under test

Link the receiver module under test to the receiving host of the test platform according to the interface specification, and then use the network transmission to connect the output data of the MIMO channel module to the receiver module under test, and the receiver module under test performs the received data Channel estimation, equalization, demodulation, and channel decoding algorithms can be tested after outputting the bit stream;

7) Use virtual instruments to display the results

According to the periodogram method, the maximum entropy spectrum estimation method or the Pisarenko harmonic decomposition method, the power spectrum estimation of the output data of the transmitter module under test is carried out, and the bit stream output by the receiver module under test is compared with the source bit stream, and the statistical total The number of error bits, and then get the bit error rate, and finally use the various display functions provided in Visual C++ to display it.

The present invention can evaluate the common wireless communication transmission technology and the transmission technology of the multi-input multi-output communication system which is very important in the future wireless communication, and connect each processor of the test platform by using the TCP/IP network protocol through the socket, Distribute the operation of the entire test to two computers, on the one hand, the simulation test is improved in terms of simulation scale and execution speed; on the other hand, the transmitter module under test and the receiver module under test are physically separated And the data received by the receiver module under test is completely determined by our third-party evaluation unit, which completely eliminates the situation that the test party short-circuits the transmitter module under test and the receiver module under test; therefore, the present invention can improve wireless communication The efficiency of transmission technology test and evaluation ensures the fairness of the test.

Description of drawings

Fig. 1 is a schematic diagram of module composition of the present invention;

Fig. 2 is the basic configuration diagram of the data exchange interface of the present invention;

Fig. 3 is a schematic diagram of the control flow of the present invention;

Fig. 4 is a schematic diagram of the SISO channel implementation method in the present invention

Fig. 5 is a schematic diagram of the interface of the transmitting host of the present invention;

Fig. 6 is a schematic diagram of the interface of the receiving host in the present invention.

Detailed ways

The present invention will be described in further detail below in conjunction with the accompanying drawings.

Referring to Fig. 1, the test platform of the present invention is composed of two processors connected through a network, that is, a transmitting host and a receiving host. Among them, the transmitting host implements the functions of the signal source, transmitting module, channel and unity detection module, while the receiving host implements the functions of the receiving module and the statistics module. In addition, both hosts are equipped with virtual instruments to observe various statistical indicators. For example: the virtual instrument of the receiving host can observe the bit error rate curve and constellation diagram. The virtual instrument of the transmitting host can observe the spectrum utilization rate, constellation diagram and power histogram of the module under test. The unity detection module is a module for unifying the duplex modes of TDD (Time Division Duplex) and FDD (Frequency Division Duplex). This facilitates a fair evaluation of the duplex modes of the two different formats. The platform of the present invention should have a variable-rate signal source generation module, a MIMO channel module, a network transmission module responsible for linking two hosts, and a virtual instrument module. Write these modules using Microsoft Visual C++:

The source bit stream with a variable rate is actually a random signal, and the random signal to be generated in the present invention is evenly distributed on two points of 0 and 1, that is, the probability of taking 0 or taking 1 is 0.5. Theoretically speaking, as long as there is a random signal of continuous distribution, the random signal of arbitrary distribution can be obtained by the corresponding transformation method, so in the simulation method of the present invention, at first produce the random signal that obeys uniformity on the (0,1) interval The random signal of distribution, then utilize transformation method to produce complex Gaussian signal, because the limitation of computer word length, the present invention can't obtain true random signal, but, as long as can pass various statistical tests of random signal, just can regard it as True random signals to use. Among the mathematical methods for generating (0, 1) uniformly distributed random signals, there are iterative centering method, shift method, multiplication congruence method and mixed congruence method. The present invention adopts the mixed congruence method with better statistical properties and wider use;

Link the transmitter module under test to the transmitter host of the test platform according to the interface specification, and then connect the generated source bit stream to the transmitter module under test, and the transmitter module under test performs channel coding, interleaving, and modulation on the received data , data output by space-time coding and framing algorithms;

For non-physical MIMO channel model simulation, the channel fading coefficient of each path of N×M subchannels in the MIMO channel is equal to the channel fading coefficient of mutually independent N×M SISO (Single Input Single Output) channels. Therefore, the simulation of the MIMO channel can be divided into two parts: 1) Generate an independent multipath fading channel. How to generate an independent multipath fading channel is a prerequisite for MIMO channel simulation. The channel fading coefficient of each path of N×M subchannels in a MIMO channel is equal to the mutually independent N×M SISO (Single-input-single-output) The product of the channel fading coefficient of the channel and the spatial correlation matrix corresponding to each path. Only after the statistically independent fading channels are generated, the MIMO channel coefficients obtained at this time can really have the spatial correlation under the specified angular power spectral density and angular expansion. Independent multipath fading channels will be affected by both time selective fading and frequency selective fading. Therefore, both frequency selective fading and time selective fading must be simulated for independent multipath fading channels. In fact, frequency selective fading is realized by treating a "broadband" signal through a multipath channel as a "narrowband" signal through multiple independent single-path channels. The time-selective fading is to use the shaping filter method to generate the Doppler power spectrum shape, the amplitude obeys the Rayleigh distribution, and the phase obeys the uniform distribution random process sampling to multiply the result of the input random signal, which is the required Time selective fading; 2) Independence verification of the generated independent multipath fading channels. Assuming that the number of paths of any one of the N×M single-input single-output channels is K, in this way, the MIMO channel can be regarded as composed of M×N×K independent single-path channels. The multipath fading channel parameters can be generated by white noise excitation. For different propagation paths, different noise seeds with good cross-correlation can be selected to ensure the uncorrelation between fading channels. To verify the uncorrelation between channels, randomly sample M×N×K independent single-path channels and select any two groups of them to form the corresponding two groups of random variables, and use the correlation function or cross-covariance function to test the two channels Independence between the two channels, if the value of the two correlation functions is zero, the two channels are independent of each other, otherwise the MIMO channel module is regenerated; using the method of hypothesis testing, we found from more than 4 billion seeds 400 seeds with guaranteed independence, among which 8×8×6=384 seeds are used to generate these 64 independent 6-path channels.

Network transmission: the present invention uses socket to realize network module, what adopt is client/server model to realize the interconnection between each host computer, and client/server model is the most commonly used paradigm of distributed application program;

Link the receiver module under test to the receiving platform of the test platform according to the interface specification, and then use network transmission to connect the output data of the MIMO channel module to the receiver module under test, and the receiver module under test performs the received data Channel estimation, equalization, demodulation, and channel decoding algorithms can be tested after outputting the bit stream;

According to the periodogram method, the maximum entropy spectrum estimation method or the Pisarenko harmonic decomposition method, the power spectrum estimation of the output data of the transmitter module under test is carried out, and the bit stream output by the receiver module under test is compared with the source bit stream to obtain The bit error rate is finally displayed by various display functions provided in Visual C++.

Referring to Fig. 2, the main control program of the test platform of the present invention provides 5 public data exchange areas, represented by A, B, C, D, E respectively, used for the data transfer between the former level module and the latter level module , or for a virtual instrument to display its output data. That is to say, the buffer is the output data of the previous level module and the input data of the next level module. Each data exchange area is a continuous storage space allocated by the system. Among them, the three data exchange areas A, B, and C are opened by the transmitting host, and the data in the C exchange area is transmitted to the D exchange area of the receiving host through two-machine communication as the input of the receiving module, that is, in the two exchange areas C and D The data is exactly the same (the simulation granularity of the two exchange areas of B and C is the same).

Referring to FIG. 3 , the entire test platform of the present invention mainly includes two types of processes: the first type is a two-computer communication process, and the second type is a virtual instrument control process. In the first process, after the data output by the channel is stored in the buffer, the read-write flag (global variable) is set to be readable, and this event is notified to the user interface thread as a message, and the user interface thread judges this event. After the message attribute, the network communication module is called to send the operation. After the network communication module receiving the host user interface thread receives the data, it sets the read and write flag (global variable) as readable to inform the worker thread that it can read from the buffer. Read the data and act accordingly. The second type of workflow is to use the critical section method to realize the thread synchronization between the virtual instrument and the working thread. First, put the source bit stream data generated by the source into buffer A. Then, the transmitter module under test provided by the party under test takes out the source bit stream data from buffer A, and puts it into buffer B after processing. The MIMO channel module processes the data in the buffer B, puts it into the C buffer, and then calls the network transmission to send the data in the C buffer to the computer on the receiving host side through the LAN. The computer on the receiving host side stores the received data into the buffer D, and the receiver module under test takes out the data from the buffer D, and after processing the data, sends the output bit stream into the buffer E . The virtual instrument reads the data from the buffer E and compares it with the source bit stream to obtain the bit error performance and spectrum utilization of the tested party. After the above operations are completed, the receiving host sends control information to the transmitting host, indicating that the current data processing is completed and the data can be sent continuously. In the above process, since the user needs to observe the waveform and power histogram of each buffer data in real time, we make the user interface thread and the worker thread share these 5 buffers, namely A, B, C, D, E . The method of thread communication between the two threads can alternately control each buffer.

Referring to FIG. 4 , the realization of the frequency selective fading characteristic of the SISO channel is the data block to be processed, that is, the data output by the transmitter module under test. After the independent fading of each path, the result is weighted in the data exchange area of the receiving end, that is, the RX position in Figure 2, that is, a "broadband" signal passing through a multipath channel can be regarded as a "narrowband" signal passing through multiple independent K Rayleigh flat fading modules to realize (assuming that a SISO channel is composed of K paths). The realization of time-selective fading of SISO channel corresponds to the Rayeigh flat fading module in Fig. 4, which uses the shaping filter method to generate the power spectral density conforming to the Jakes model, the amplitude obeys the Rayleigh distribution, and the phase obeys the uniform distribution stochastic process.

Referring to Figure 5, the two dark areas on the upper side of the figure show the spectrum utilization measured by the tested party's wireless transmission technology solution test platform under specific channel conditions.

Referring to Figure 6, the two dark areas on the upper side of the figure show the bit error rate performance measured by the test platform under specific channel conditions of the wireless transmission technology solution of the tested party.

Claims (1)

1. the method for designing of distributed wireless communication transmission technique test platform is characterized in that:

1) generation of source bits stream

At first utilize iteration to get the random signal train that a binary sequence that middle method, displacement method, multiplicative congruential method or mixed congruence method will import at random produces its conversion a compound multiple Gaussian Profile, and then be created on (0,1) interval by the hypothesis test of random signal and obey equally distributed and can be at the random sources bit stream of freely regulated speed, frame length between the 8Kbps to 100Mbps;

2) link tested transmitter module

Tested transmitter module chain is tapped into the emission main frame of test platform according to interface specification, then the source bits stream that generates is inserted tested transmitter module, tested transmitter module to the data that receive carry out chnnel coding, interweave, the data of modulation, Space Time Coding and the output of framing algorithm;

3) generation of multiple-input and multiple-output mimo channel module

Suppose that the multiple-input and multiple-output mimo channel has M transmit antennas and N reception antenna, that is M input and N export, mimo channel just is equivalent to the independently single single output of input SISO channel of M * N bar so, each is independently singly imported single output SISO channel and has frequency selective fading and time selective fading two specific characters, the realization of frequency selective fading characteristic is that " broadband " signal is become " arrowband " signal by multipath channel, thereby obtain required frequency selective fading by a plurality of independently single footpaths channel weights then, the realization of time selective fading characteristic is a method of utilizing forming filter, generation meets Doppler's power spectrum shape, the whose amplitude obeys rayleigh distributed, phase place is obeyed equally distributed random process sampled signal and is gone to multiply by the result that the output signal of tested transmitter obtains, thereby obtain required time selective fading, promptly generate the M * single output of N bar list input SISO channel, and then obtained the multiple-input and multiple-output mimo channel;

4) checking of multiple-input and multiple-output mimo channel module independence

The footpath number of supposing the multipath channel between any in the M * single output of N bar list input SISO channel is K, M * N * K independently single footpath channel carried out stochastical sampling get wherein any two groups of two groups of stochastic variables that constitute correspondence, check the independence of two interchannels with correlation function or cross covariance function, if the value of two correlation functions is that zero two channels are separate, otherwise the generation that changes multiple-input and multiple-output mimo channel module over to regenerates multiple-input and multiple-output mimo channel module;

5) Network Transmission of data

Tested transmitter module behind mimo channel, utilizes socket to realize the ICP/IP protocol Network Transmission data of output, is about to these data and is sent to the reception main frame, reads this data by the tested receiver module that receives in the main frame;

6) link tested receiver module

The tested receiver Module Links is advanced the reception main frame of test platform according to interface specification, utilize Network Transmission that the dateout of multiple-input and multiple-output mimo channel module is inserted the tested receiver module then, the tested receiver module carries out the data that receive can testing after channel estimating, equilibrium, demodulation, the channel decoding algorithm output bit flow;

7) utilize the virtual instrument display result

According to period map method, Maximum Entropy Spectral Estimation method or Pisarenko harmonic wave decomposition method are carried out power spectrum estimation to the dateout of tested transmitter module and the bit stream and the source bits stream of tested receiver module output are contrasted, add up total error bit number, and then obtain the error rate, utilize the various demonstration class functions that provide in the visual c++ to show at last.

Images