LTE维也纳大学仿真器说明文档.pdf

发布时间：2022-06-13 发布人：admin 分类：说明书资料大小：2.80M 资料格式：pdf 举报版权申诉

1222e0ee-553c-4ebc-84fc-899889c2b923.pdf-第1页.png

第1页 / 共14页

1222e0ee-553c-4ebc-84fc-899889c2b923.pdf-第2页.png

第2页 / 共14页

1222e0ee-553c-4ebc-84fc-899889c2b923.pdf-第3页.png

第3页 / 共14页

1222e0ee-553c-4ebc-84fc-899889c2b923.pdf-第4页.png

第4页 / 共14页

1222e0ee-553c-4ebc-84fc-899889c2b923.pdf-第5页.png

第5页 / 共14页

1222e0ee-553c-4ebc-84fc-899889c2b923.pdf-第6页.png

第6页 / 共14页

1222e0ee-553c-4ebc-84fc-899889c2b923.pdf-第7页.png

第7页 / 共14页

1222e0ee-553c-4ebc-84fc-899889c2b923.pdf-第8页.png

第8页 / 共14页

I Foreword

II Running the simulator for the first time

III MEX files

IV Running your own simulations

V Simulation parameters

V-A General parameters

V-B Channel matrix source

V-C UE, eNodeB and UE feedback specific parameters

V-D Channel model parameters

V-E Scheduler parameters

VI Reference simulation results included

VII Reproducing the results/plots presented on a paper

VIII SNR definition

IX Note on parallel simulations and random number generation

X Changelog

XI Referencing

XII Known issues

XIII Using the Winner Phase II channel model reference implementation

XIV Questions

XV Mailing List

XVI The People (so far) behind the development of the simulator

XVII License agreement

XVII-A Academic Usage

XVII-B Grant of copyright license

XVII-C Grant of source code license

XVII-D Exclusions from license grant

XVII-E Warranty of provenance and disclaimer of warranty

XVII-F Limitation of liability

XVII-G Termination

XVII-H Open source and code under other license terms

XVII-I Appendix I

References

Vienna LTE Simulators Link Level Simulator Documentation, v1.7r1089 1 Institute of Telecommunications Vienna University of Technology, Austria Gusshausstrasse 25/389, A-1040 Vienna, Austria Email: {jcolom, msimko, sschwarz}@nt.tuwien.ac.at Web: http://www.nt.tuwien.ac.at/ltesimulator This document contains documentation on how to use the Long Term Evolution (LTE) Link Level simulator [1] from the Vienna LTE simulator suite as well as some insight on its structure and the assumptions that were made while developing it. This document relates more on how to actually use the simulator. The concept and the structure of the simulator is described in more detail in [2]. Abstract I. FOREWORD The LTE link level simulator is published under a non-commercial academic use license. Please make sure that you understand the terms and conditions of the license before you use any of the available software packages. Would you require a license different to a non-commercial academic one please contact Markus Rupp, or Josep Colom Ikuno. The detailed license agreement for the LTE Link Level simulator can be found in Section XVII. Please read the license agreement carefully, as parts of the code are under the GNU Lesser General Public License [3], and the MIT License [4]. II. RUNNING THE SIMULATOR FOR THE FIRST TIME The LTE Link Level simulator simulator is provided with a test simulation scenario that can be used to verify that the simulator runs in the expected way. In the main directory you can execute the LTE_sim_batch_quick_test MATLAB script. This script provides a means to test the performance of an LTE transmission on an uncorrelated PedB channel and ﬂat Rayleigh channel for several transmission modes [5]. Since currently the feedback calculation for the Closed Loop Spatial Multiplexing (CLSM) mode is not implemented, only the Single-Input Single-Output (SISO), Transmission Diversity (TxD) and Open Loop Spatial Multiplexing (OLSM) modes are simulated. The result of the simulation is a group of result ﬁles that is stored in the ./results folder. The plots_quick_test_results MATLAB script provides the means of plotting the results for this speciﬁc simulation. By executing the script without any change, the Block Error Ratio (BLER) and throughput results for the following simulations are shown: • 5000 Transmission Time Interval (TTI) long, ﬂat Rayleigh channel, no retransmissions: Figures 1(a) and 1(c). • 5000 TTI long, ﬂat Rayleigh channel, maximum of three retransmissions: Figures 1(b) and 1(d). • 5000 TTI long, uncorrelated PedB channel, no retransmissions: Figures 2(a) and 2(c). • 5000 TTI long, uncorrelated PedB channel, maximum of three retransmissions: Figures 2(b) and 2(d). The ﬁgures obtained by running LTE_sim_batch_quick_test and the ones shown in Figures 1 and 2 should match. The settings used in the LTE_sim_batch_quick_test MATLAB script can be found in Table I, although a more detailed description can be found in Section V and the LTE_load_parameters_SUSISO_quick_test (SISO mode) and LTE_load_parameters_SUMIMO_quick_test (MIMO modes) conﬁgurations ﬁles. III. MEX FILES Computation-intensive parts of the LTE link level simulator are implemented in C and used in the simulator by means of MEX ﬁles. The source code for the given ﬁles can be found in the /C-source folder under the simulator root folder. Please not that some of the functions there are licensed under other license terms. Please check Section XVII for more details. The MEX ﬁles distributed with the simulator release are the Windows 64-bit version. Should you require them for any other platform you recompile them by using the LTE_aux_mex_files script. You can ﬁnd more information on how to use/write/compile MEX ﬁles here.

2 (a) BLER, ﬂat Rayleigh channel, no HARQ (b) BLER, ﬂat Rayleigh channel, 3 retransmissions (c) Throughput, ﬂat Rayleigh channel, no HARQ (d) Throughput, ﬂat Rayleigh channel, 3 retransmissions Fig. 1. Plots resulting from running the LTE_sim_batch_quick_test.m MATLAB script (ﬂat rayleigh) Parameter Number of User Equipments (UEs) Bandwidth Retransmissions Channel type Filtering Receiver type Simulation length Transmit modes Value 1 1.4 MHz 0 and 3 Flat Rayleigh, PedB uncorrelated Block Fading Soft Sphere Decoder 5000 subframes SISO, TxD (2x1 and 4x2) and OLSM (4x2) BASIC SETTINGS USED FOR THE LTE_S I M_B A T C H_Q U I C K_T E S T.M MATLAB SCRIPT TABLE I IV. RUNNING YOUR OWN SIMULATIONS While the LTE_sim_batch_quick_test is adequate to get familiar with the basic structure of the simulation script necessary to run it, you may need more ﬂexibility than the one offered with the quick test script. For this, you can check the LTE_sim_batch.m script. It provides the means to run a simulation for a speciﬁc set of Modulation and Coding Schemes (MCSs). Below you can ﬁnd a list of exemplary parameters that you may want to conﬁgure in the batch ﬁle: • cqi_i: set of MCSs that are used for the simulation. In [6], 15 different Channel Quality Indicators (CQIs) are speciﬁed, denominated CQIs. If you want to simulate for all possible CQIs, just set cqi_i to be [1:15]. Analogously, in the MATLAB notation, for a simulation for CQIs 1 and 7 that would be [1 7]. −10−50510152010−310−210−1100BLER, CQI 7, PedB, 5000 subframes, at rayleigh, 0 re−txBLERSNR [dB] SISOTxD 2x1TxD 4x2OLSM 4x2−10−50510152010−310−210−1100BLER, CQI 7, PedB, 5000 subframes, at rayleigh, 3 re−txBLERSNR [dB] SISOTxD 2x1TxD 4x2OLSM 4x2−10−50510152000.511.522.5throughput, CQI 7, PedB, 5000 subframes, at rayleigh, 0 re−txthroughput [Mbps]SNR [dB] SISOTxD 2x1TxD 4x2OLSM 4x2−10−50510152000.511.522.5throughput, CQI 7, PedB, 5000 subframes, at rayleigh, 3 re−txthroughput [Mbps]SNR [dB] SISOTxD 2x1TxD 4x2OLSM 4x2

3 (a) BLER, PedB channel, no HARQ (b) BLER, PedB channel, 3 retransmissions (c) Throughput, PedB channel, no HARQ (d) Throughput, PedB channel, 3 retransmissions Fig. 2. Plots resulting from running the LTE_sim_batch_quick_test.m MATLAB script (PedB) • N_subframes: the length of the simulation, or how many subframes (TTIs) are simulated. • SNR_vec: a vector containing the Signal to Noise Ratios (SNRs) that will be used for each simulation run. Use an SNR range adequate to the CQI that you are simulating. The vectors set in the script are adequate for a SISO Additive White Gaussian Noise (AWGN) simulation. See Section VIII for a more detailed description of how SNR is deﬁned. • LTE_load_parameters: load the parameter ﬁle that conﬁgures the simulator. A more detailed description of the available conﬁguration parameters can be found in Section V. Four basic preconﬁgured options are given that can be directly used by appropriately setting the LTE_params.Simulation_type variable accordingly, thus skipping the conﬁguration of the rest of the parameters. See Section VII if you want to use the parameter ﬁle to reproduce the results/plots from a paper. – SUSISO: Single-user SISO simulation – MUSISO: Multi-user (preconﬁgured to two users) SISO simulation – SUMIMO: Single-user Multiple-Input Multiple-Output (MIMO) simulation – MUMIMO: Multi-user (two users) MIMO simulation • LTE_sim_main: main routine of the simulator, also called by the LTE_sim_batch_XXX ﬁles. • Generate the output ﬁlename and save the results in a .mat ﬁle. Below you can ﬁnd a list of the parameters that can be conﬁgured in the LTE_config ﬁle: V. SIMULATION PARAMETERS −10−50510152010−310−210−1100BLER, CQI 7, PedB, 5000 subframes, PedB, 0 re−txBLERSNR [dB] SISOTxD 2x1TxD 4x2OLSM 4x2−10−50510152010−310−210−1100BLER, CQI 7, PedB, 5000 subframes, PedB, 3 re−txBLERSNR [dB] SISOTxD 2x1TxD 4x2OLSM 4x2−10−50510152000.511.522.5throughput, CQI 7, PedB, 5000 subframes, PedB, 0 re−txthroughput [Mbps]SNR [dB] SISOTxD 2x1TxD 4x2OLSM 4x2−10−50510152000.511.522.5throughput, CQI 7, PedB, 5000 subframes, PedB, 3 re−txthroughput [Mbps]SNR [dB] SISOTxD 2x1TxD 4x2OLSM 4x2

A. General parameters 4 • LTE_params.nUE: number of UEs to simulate. • LTE_params.nBS: number of eNodeBs (cells) that will be simulated. Do not change this parameter, as support for multiple eNodeBs is not yet implemented. • LTE_params.uplink_delay: the delay the uplink experiences. It applies to ACKnowledgments (ACKs), CQI, Precoding Matrix Indicator (PMI) and Rank Indicator (RI) reports. An integer number ≥ 0. Useful if you want to experiment with scheduling algorithms or feedback strategies. • LTE_params.show_plots: whether plots are shown during the simulation or not. • LTE_params.trace_subcarrier_SNR: if set to true, a trace of the subcarrier SNRs is generated and stored in the simulation_results object. • LTE_params.N_seed_reset: resets the random number generator seeds to a new value after LTE_params.N_seed_reset subframes. This is used for the case where a time-correlated channel is generated (modiﬁed Rosa Zheng model [7], [8]). When low speed channels are evaluated, a too-long simulation would be needed in order to obtain statistically meaningful results. Thus, to avoid such situations, the channel can be ”reset” to a new seed every N subframes. • LTE_params.carrier_freq: carrier center frequency [Hz] • LTE_params.Bandwidth: system bandwidth. Allowed values are 1.4 MHz, 3 MHz, 5 MHz, 10 MHz, 15 MHz, and 20 MHz. This bandwidths are equivalent to 6, 15, 25, 50, 75, and 100 Resource Blocks (RBs) respectively. • LTE_params.HARQ_processes: number of parallel Hybrid-ARQ (HARQ) processes. The maximum value, according to [9] is 8. • LTE_params.max_HARQ_retransmissions: maximum number of HARQ retransmissions, not including the orig- inal transmission. Valid values are 0, 1, 2 or 3. Higher numbers will give you an error, as the rate matching is not deﬁned for retransmission numbers (rv_idx) higher than 3. • LTE_params.SubcarrierSpacing: in Hz, 15 kHz, a 7.5 kHz subcarrier spacing is also possible (just for MBSFN- based multicast/broadcast transmissions). Tests were so far performed using a 15 kHz spacing, so the 7.5 kHz spacing is not thoroughly tested. • LTE_params.CyclicPrefix: cyclic preﬁx length [10]. Either normal or extended for MBSFN-based multi- cast/broadcast transmissions. • LTE_params.simulation_type: the simulator is capable of using the MATLAB Parallel Toolbox in order to speed up simulations by using parfor loops. If you happen to have the Distributed Computing Toolbox, you will also be able to make use it by using this option. Set this variable to parallel or normal to parallelize the SNR loop in LTE_sim_main or just perform a single-core simulation. Keep in mind that some modiﬁcations you do to the code may not work in the parallel version or may directly cause it not to run. • LTE_params.simulate_with_all_zero_sequences: true if you want that the transmitted data is an all-zero sequence (useful for interleaver testing). • LTE_params.introduce_frequency_offset: whether you want a frequency offset to be introduced. Frequency offset introduction and correction are currently under development, so for now this option may not yet be fully functional. Further carrier offset options are also present in the UE parameters conﬁguration Section V-C. • LTE_params.random_noise_seeding: whether the seed for the random number generator that generates the noise is set (allows for repeatability of the noise realizations). • LTE_params.noise_seed: Only used if the upper variable is set to true. Integer number that sets the random number seed of the noise random number generator. • LTE_params.usePBCH: whether space for the physical broadcast channel is reserved inside the resource grid (no data is transmitted on these resource elements). • LTE_params.usePDCCH: whether space for the physical downlink control channel is reserved inside the resource grid. • LTE_params.trafficmodel.usetraffic_model: whether users generate trafﬁc according to prespeciﬁed trafﬁc models (RAN R1-070674) or a full buffer situation is assumed. Currently the trafﬁc models are not used in the simulations (only full buffer is possible). B. Channel matrix source • LTE_params.channel_matrix_source: Controls the generation of the channel matrix trace. generated to generate it every time. trace to load it from a trace. • LTE_params.store_channel_trace: Set to true or false. If mode is generated, the channel trace will be saved at the end of the simulation. • LTE_params.channel_matrix_tracefile: ﬁlename of the trace ﬁle where the generated channel matrix trace is stored. Only applicable if trace mode is used (if the mode is set to trace, the channel matrix is already read from a trace, so it is meaningless to save it again in another trace).

C. UE, eNodeB and UE feedback speciﬁc parameters 5 • LTE_params.UE_config.LLR_clipping: speciﬁes the LLR clipping level. • LTE_params.UE_config.turbo_iterations: Number of iterations of the turbo decoder. Set by default to 8. • LTE_params.UE_config.N_soft: Deﬁnes the total number of soft channel bits available for HARQ processing (TS • LTE_params.UE_config.channel_estimation_method: Currently the following channel estimators are avail- • LTE_params.UE_config.channel_interpolation_method: linear, cubic, spline, sinc_freq, sinc_time, or T-F. For fast fading, linear, cubic, and v4 are available. • LTE_params.UE_config.autocorrelation_matrix_type: type of autocorrelation matrix. Either ideal or 36.306 4.2.1.3 [11]). able: PERFECT, LS, or MMSE. estimated. channel autocorrelation matrix. • LTE_params.UE_config.user_speed: channel speed (m/s). • LTE_params.UE_config.realization_num: number of channel realizations. Used for averaging to obtain the • LTE_params.UE_config.realization_num_total: ﬁrst xy number of channel realizations are used just for the estimation of the autocorrelation matrix. • LTE_params.UE_config.CDD: Cyclic Delay diversity – 0: zero delay CDD (3GPP TS 36.211-820 Section 6.3.4.2.1 [10], page 37) – 1: small delay CDD (3GPP TS 36.211-820 Section 6.3.4.2.1 [10], page 37). Although in the newest standard version this is not deﬁned anymore – 2: large delay CDD (3GPP TS 36.211-820 Section 6.3.4.2.2 [10], page 38) • LTE_params.UE_config.mode: the transmission modes are deﬁned in TS 36.213-820 Section 7.1, page 12 [6] and are: – 1: single antenna. – 2: TxD. – 3: OLSM. – 4: CLSM. – 5: Multiuser MIMO (not yet implemented) • LTE_params.UE_config.nRX: number of receive antennas at the UE. • LTE_params.UE_config.carrier_freq_offset: carrier frequency offset normalized to subcarrier spacing (not • LTE_params.UE_config.perfect_freq_sync: whether the UE is perfectly synchronized in frequency. • LTE_params.UE_config.rfo_correct_method: receiver frequency offset correction method. Either none, subframe, yet implemented). or FIR. Not yet implemented. • LTE_params.UE_config.receiver: either SSD for a Soft Sphere Decoder, ZF for a Zero Forcing (ZF) receiver and MMSE for a Minimum Mean Square Error (MMSE) receiver. • LTE_params.UE_config.PMI_fb_granularity: granularity of PMI feedback in multiples of resource blocks • LTE_params.UE_config.CQI_fb_granularity: granularity of CQI feedback in multiples of resource blocks • LTE_params.UE_config.PMI_fb: whether PMI feedback is activated or not (true/false) • LTE_params.UE_config.RIandPMI_fb: whether RI feedback is activated in addition to PMI feedback (true/false) • LTE_params.UE_config.CQI_fb: whether CQI feedback is activated • LTE_params.UE_config.predict: whether channel prediction is activated for the feedback calculation; the channel prediction algorithm is simple linear extrapolation on every subcarrier; • LTE_params.UE_config.SINR_averaging.averager: deﬁnes the Effective Signal to Interference and Noise Ratio Mapping (ESM) averager used. Possible values are EESM and MIESM. • LTE_params.UE_config.SINR_averaging.EESMbetas: deﬁnes the calibration factors required for Exponential Effective Signal to Interference and Noise Ratio Mapping (EESM). Those values are obtained from extensive training simulations and should not be changed. • LTE_params.UE_config.SINR_averaging.MIESMbetas: deﬁnes the calibration factors required for Mutual Information Effective Signal to Interference and Noise Ratio Mapping (MIESM). Those values are obtained from extensive training simulations and should not be changed. • LTE_params.UE_config.SINR_averaging.MCSs: deﬁnes the used MCS set. Modiﬁcation of these values might lead to unexpected behaviour, as the calibration factors (EESMbetas and MIESMbetas) are not trained for other MCSs. • LTE_params.BS_config.nTx: number of antennas at the eNodeB. • LTE_params.feedback.ignore_channel_estimation: whether the channel estimation mean square error is taken into account during the feedback calculation (as additional noise) or not • LTE_params.feedback.channel_averaging: whether channel averaging is used during feedback calculation or

not. If set to true just a single average channel value per resource block is used to compute the feedback for complexity reduction. Especially in 4 × 4 systems this degrades the performance of the feedback method. D. Channel model parameters 6 Parameters that conﬁgure how the channel is generated and the signal ﬁltered. • LTE_params.ChanMod_config.filtering: BlockFading (channel is constant during one subframe) or FastFading. • LTE_params.ChanMod_config.interpolation_method: the channel interpolation method for the channels which are generated in the simulator. Either shift_to_nearest_neighbor for nearest neighbor interpolation or sinc_interpolation for sinc interpolation, which is more precise. Necessary if the channel sampling rate is not equal to the sampling rate of the transmit signal. • LTE_params.ChanMod_config.sin_num: speciﬁes the number of sin realizations used for the modiﬁed rosa-zheng model [7], [8]. • LTE_params.ChanMod_config.type: speciﬁes the type of channel used. The available ones are: wavelength ers (yes or no). grid for all links (yes or no). 7=B5a, 8=B5c, 9=B5f, 10=C1, 11=C2, 12=C3, 13=C4, 14=D1 and 15=D2a. – AWGN: Additive White Gaussian Noise channel. – flat Rayleigh: ﬂat Rayleight channel. – Tap-delay based models: PedA, PedB, PedBcorr, VehA, VehB, TU, RA, and HT [12], [13]. – Externally-generated channel coefﬁcients: winner_II. Uses the publicly-availabe Winner II implementation to generate the channel coefﬁcients [14]. The following parameters can be conﬁgured when using the Winner II channel model. ∗ LTE_params.ChanMod_config.winner_settings.Scenario: 1=A1, 2=A2, 3=B1, 4=B2, 5=B3, 6=B4, ∗ LTE_params.ChanMod_config.winner_settings.PropagCondition: LOS or NLOS. ∗ LTE_params.ChanMod_config.winner_settings.SampleDensity: number of time samples per half ∗ LTE_params.ChanMod_config.winner_settings.UniformTimeSampling: use same time sampling ∗ LTE_params.ChanMod_config.winner_settings.FixedPdpUsed: nonrandom path delays and pow- ∗ LTE_params.ChanMod_config.winner_settings.FixedAnglesUsed: nonrandom AoD/AoAs (yes ∗ LTE_params.ChanMod_config.winner_settings.PolarisedArrays: usage of dual polarised arrays ∗ LTE_params.ChanMod_config.winner_settings.TimeEvolution: usage of time evolution (yes or ∗ LTE_params.ChanMod_config.winner_settings.PathLossModelUsed: usage of path loss model ∗ LTE_params.ChanMod_config.winner_settings.ShadowingModelUsed: usage of shadow fading ∗ LTE_params.ChanMod_config.winner_settings.PathLossModel: path loss model function name ∗ LTE_params.ChanMod_config.winner_settings.PathLossOption: Available options are CR_light, ∗ LTE_params.ChanMod_config.winner_settings.RandomSeed: sets a random seed. Can be left empty. ∗ LTE_params.ChanMod_config.winner_settings.UseManualPropCondition: whether to use man- ual propagation condition (LOS/NLOS) settings or not (yes or no). If not, the propagation condition is drawn from probabilities. CR_heavy, RR_light, RR_heavy. CR=Corridor-Room, RR=Room-Room NLOS. model (yes or no). or no). (yes or no). no). (yes or no). (pathloss). • LTE_params.ChanMod_config.corr_coefRX: correlation between the receiver antennas. Only compatible with • LTE_params.ChanMod_config.corr_coefTX: correlation between the transmitter antennas. Only compatible with block fading ﬁltering. block fading ﬁltering. • LTE_params.ChanMod_config.time_correlation: sets whether the channel realizations are time-correlated or not. correlated or independent. This parameter is also related with LTE_params.N_seed_reset. E. Scheduler parameters Conﬁguration of the scheduler parameters. • LTE_params.scheduler.type: type of scheduler. Either round robin, best cqi, or fixed.

7 • LTE_params.scheduler.assignment: either static, semi static or dynamic. Whether the scheduler will statically assign or dynamically assign CQIs and other params. The best cqi scheduler is capable of assigning MCSs dynamically. The semi static scheduler adapts automatically the precoder and layer number according to the PMI and RI feedback (use this one in conjunction with closed loop spatial multiplexing and activated PMI and RI feedback). The semi static scheduler assigns resource blocks in a round robin fashion. • LTE_params.scheduler.fixed_scheduler_assignment: this option is used only for the fixed scheduler. In that case, the parameter is a vector of length LTE_params.nUE containing the number of RBs allocated to each user. The total number or assigned RBs must be equal or less than the number of RB available in the speciﬁed bandwidth: eg. for 2 UEs and 1.4 MHz [4 2] would assign four RBs to the ﬁrst user and two to the second. • LTE_params.scheduler.cqi: the CQI the scheduler is going to use when transmitting data. When set to ’set’, when in the static case, the CQI value to use will be read from the cqi_i variable, which is to be set from the script ﬁle that launches the simulation (eg. LTE_sim_batch). • LTE_params.scheduler.PMI: sets the Precoding Matrix Indicator feedback for the CLSM transmit mode. VI. REFERENCE SIMULATION RESULTS INCLUDED Besides the one mentioned on Section II, the simulator is provided with some reference simulation results, which can be compared with performance curves from 3GPP RAN documents such as [15] to cross-check the results of the simulator. The example simulation results distributed with the simulator are described below. • plot_reference_BLER_curves_r548 shows the plots in Figure 3, showing BLER and throughput for an AWGN simulation using the MCS deﬁned in [6] (CQIs 1-15). • plot_R1071967_throughput_curves_r553 shows the BLER and throughput curves obtained from performing AWGN simulations with the MCSs speciﬁed in R1-071967, page 16 [15], as seen in Figure 4. The MCS, including the CQI value used in the simulator, the modulation and Effective Code Rate (ECR) used in each of the simulations are shown in Table II. CQI Modulation 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 4QAM 4QAM 4QAM 4QAM 4QAM 4QAM 4QAM 4QAM 4QAM 4QAM 16QAM 16QAM 16QAM 16QAM 16QAM 16QAM 16QAM 16QAM 16QAM 64QAM 64QAM 64QAM 64QAM 64QAM 64QAM 64QAM 64QAM ECR [15] 1/9 1/6 0.21 1/4 1/3 0.42 1/2 0.58 2/3 0.73 0.43 0.46 1/2 0.54 0.58 0.61 2/3 0.73 4/5 0.58 0.62 2/3 0.70 0.74 4/5 0.85 0.90 ECRx1024 114 171 215 256 314 430 512 594 683 748 440 471 512 553 594 625 683 748 819 594 635 683 717 758 819 870 922 CQI Modulation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 4QAM 4QAM 4QAM 4QAM 4QAM 4QAM 16QAM 16QAM 16QAM 64QAM 64QAM 64QAM 64QAM 64QAM 64QAM ECR 0.0762 0.1172 0.1885 0.3008 0.4385 0.5879 0.3691 0.4785 0.6016 0.4551 0.5537 0.6504 0.7539 0.8525 0.9258 ECRx1024 [6] 78 120 193 308 449 602 378 490 616 466 567 666 772 873 948 MCSS USED IN THE LTE_S I M_B A T C H_R1_07196 (LEFT) AND LTE_S I M_B A T C H_Q U I C K_T E S T (RIGHT) MATLAB SCRIPTS TABLE II

8 (a) BLER, AWGN no HARQ (b) Throughput, AWGN, no HARQ Fig. 3. Reference BLER and throughput plots for the 15 MCSs deﬁned in [6] (a) BLER, AWGN no HARQ (b) Throughput, AWGN, no HARQ Fig. 4. Reference BLER and throughput plots for the 27 MCSs deﬁned in [15] VII. REPRODUCING THE RESULTS/PLOTS PRESENTED ON A PAPER One of the main points of the simulator is to allow you to reproduce and review our results, as well as the algorithms that produce them. For each of the following publications a script is prepared in the folder paper scripts that will reproduce the results in the corresponding paper: • M. ˇSimko, C. Mehlf¨uhrer, M. Wrulich, and M. Rupp, “Doubly Dispersive Channel Estimation with Scalable Complexity”, in Proc. WSA 2010, Bremen, Germany, Feb. 2010 [16]: LTE_sim_batch_michal_wsa_2010.m’ • S. Schwarz, M. Wrulich, and M. Rupp, “Mutual information based calculation of the precoding matrix indicator for 3GPP UMTS/LTE”, in Proc. WSA 2010, Bremen, Germany, Feb. 2010 [17]: LTE_sim_batch_schwarz_wsa_2010.m’ • S. Schwarz, C. Mehlf¨uhrer, and M. Rupp, “Calculation of the Spatial Preprocessing and Link Adaption Feedback for 3GPP UMTS/LTE”, in Proc.Wireless Advanced2010, London, Great Britain, June 2010 [18]: LTE_sim_batch_schwarz_WiAd_2010.m’ In a MIMO-Orthogonal Frequency Division Multiplexing (OFDM) transmission system the SNR γ is the measurement for channel quality information and is a key factor of link error prediction. There are different measures or calculation procedures VIII. SNR DEFINITION -15-10-5051015202510-310-210-1100BLER, 1.4MHz, SISO AWGN, 5000 subframesBLERSNR [dB] CQI 01CQI 02CQI 03CQI 04CQI 05CQI 06CQI 07CQI 08CQI 09CQI 10CQI 11CQI 12CQI 13CQI 14CQI 15-15-10-505101520250123456throughput, 1.4MHz, SISO AWGN, 5000 subframesthroughput [Mbps]SNR [dB] CQI 01CQI 02CQI 03CQI 04CQI 05CQI 06CQI 07CQI 08CQI 09CQI 10CQI 11CQI 12CQI 13CQI 14CQI 15−15−10−5051015202510−310−210−1100BLER, 1.4MHz, SISO AWGN, 5000 subframesBLERSNR [dB] QPSK, R=1/9QPSK, R=1/6QPSK, R=0.21QPSK, R=1/4QPSK, R=1/3QPSK, R=0.42QPSK, R=1/2QPSK, R=0.58QPSK, R=2/3QPSK, R=0.7316QAM, R=0.4316QAM, R=0.4616QAM, R=1/216QAM, R=0.5416QAM, R=0.5816QAM, R=0.6116QAM, R=2/316QAM, R=0.7316QAM, R=4/564QAM, R=0.5864QAM, R=0.6264QAM, R=2/364QAM, R=0.7064QAM, R=0.7464QAM, R=4/564QAM, R=0.8564QAM, R=0.90−15−10−505101520250123456throughput, 1.4MHz, SISO AWGN, 5000 subframesthroughput [Mbps]SNR [dB] QPSK, R=1/9QPSK, R=1/6QPSK, R=0.21QPSK, R=1/4QPSK, R=1/3QPSK, R=0.42QPSK, R=1/2QPSK, R=0.58QPSK, R=2/3QPSK, R=0.7316QAM, R=0.4316QAM, R=0.4616QAM, R=1/216QAM, R=0.5416QAM, R=0.5816QAM, R=0.6116QAM, R=2/316QAM, R=0.7316QAM, R=4/564QAM, R=0.5864QAM, R=0.6264QAM, R=2/364QAM, R=0.7064QAM, R=0.7464QAM, R=4/564QAM, R=0.8564QAM, R=0.90

分享到：

赞收藏

资料库

LTE维也纳大学仿真器说明文档.pdf

相关推荐

课程资源

热门标签

最新资料