Nghiên cứu và thiết kế phương pháp ước lượng kênh truyền sử dụng bộ lọc wiener cho mạng di động LTE

Nguyen Duy Phuong - CB120720 CONTENTS CONTENTS CONTENTS OF TABLES CONTENTS OF FIGURES ABBREVIATIONS ABSTRACT CHAPTER I : INTRODUCTION TO LTE 1.1 The Emergence of LTE 1.1.1 Choice of technologies for LTE 12 1.2 Thesis Objectives 13 1.2.1 Thesis Outline 13 CHAPTER II : OVERVIEW OF LTE PHY LAYER 15 2.1 Downlink Physical Channels 15 2.2 LTE Frame Structure 18 2.3 LTE Channel Models 23 2.4 Downlink Physical Signals 26 CHAPTER III: CHARACTERISTIC OF RADIO CHANNEL AND CHANNEL ESTIMATION 28 3.1 Characteristics of a Mobile Radio Channel 28 3.2 Characterization of Fading Channels 29 3.3 Orthogonal Frequency Division Multiplexing (OFDM) 32 3.4 Multi-antenna Transmission Schemes 35 3.5 Channel Estimation 36 3.5.1 Pilot Structure 37 3.5.2 Channel Estimation in an OFDM Based System 39 3.5.3 Wiener Filtering in an OFDM System 42 CHAPTER IV: SYSTEM MODEL AND WIENER FILTERING 43 4.1 System Model of a MIMO-OFDM System 43 4.2 Channel Estimation Based on Cell-specific Reference Signal 44 Page Nguyen Duy Phuong - CB120720 4.3 Wiener Based Channel Estimation 47 4.3.1 Channel Estimation in Frequency Direction 48 4.3.2 Channel Estimation in Time Direction 49 CHAPTER V: CHANNEL PARAMETERS ESTIMATION 53 5.1 Delay Spread Estimation 53 5.1.1 Delay Spread Implementation 55 5.2 Doppler Spread Estimation 56 5.2.1 Doppler spread implementation 57 5.3 SNR estimation 58 5.3.1 Implementation for the SNR estimation 58 CHAPTER VI: SIMULATION RESULTS AND ANALYSIS 60 6.1 Experimental setup 60 6.2 Simulation for the pilot channel estimates 61 6.3 Simulation for Wiener interpolation in frequency 63 6.4 MSE performance of LTE channel models 63 6.5 MSE performance for different power delay profiles 64 6.6 MSE performance for estimated delay spread 66 6.7 Doppler spread performance 67 6.8 MSE performance based on Wiener time interpolation 69 CHAPTER VII: CONSLUTION AND FUTURE WORK 72 REFERENCE 73 Page Nguyen Duy Phuong - CB120720 CONTENTS OF TABLES Table 2-1: Typical Parameters for Downlink Transmission 17 Table 2-2: Extended Pedestrian A model (EPA) 24 Table 2-3: Extended Vehicular A model (EVA) 24 Table 2-4: Extended Typical Urban model (ETU) 25 Table 2-5: E-UTRA Delay Profiles 25 Table 4-1: Comparison of cell-specific reference signal structure 46 Table 6-1: Simulation parameter settings 60 Page Nguyen Duy Phuong - CB120720 CONTENTS OF FIGURES Figure 2-1: Downlink resource grid .16 Figure 2-2: Frame Structure Type 19 Figure 2-3: Frame Structure Type 20 Figure 2-4: Overview of Physical Channel Processing 20 Figure 3-1: A typical mobile radio environment 28 Figure 3-2: Processing components in an OFDM system 34 Figure 3-3: Block-type pilot arrangement [7] 38 Figure 3-4: Comb-type pilot arrangement [7] .39 Figure 3-5: MMSE channel estimation 41 Figure 4-1: Main processing components of the MIMO-OFDM system model 43 Figure 4-2: Cell-specific reference signal: Downlink reference signal structure47 Figure 6-1: EPA channel: Plot of ideal channel vs estimated channel .62 Figure 6-2: EVA channel: Plot of ideal channel vs estimated channel 62 Figure 6-3: ETU channel: Plot of ideal channel vs estimated channel .62 Figure 6-4: Comparison of ideal channel vs Wiener interpolated channel .63 Figure 6-5: Wiener interpolation: MSE comparison for LTE channel models 63 Figure 6-6: Performance comparison based on EPA channel .64 Figure 6-7: Performance comparison based on EVA channel 64 Figure 6-8: Performance comparison based on ETU channel .65 Figure 6-9: EPA channel: actual vs estimated delay spread .66 Figure 6-10: EVA channel: actual vs estimated delay spread 66 Figure 6-11: ETU channel: actual vs estimated delay spread 67 Figure 6-12: MSE performance between actual and estimated Doppler spread68 Figure 6-13: MSE performance between actual and estimated Doppler spread68 Figure 6-14: SNR comparison: actual vs estimated SNR 69 Figure 6-15: Comparison of Wiener time filtering for different antenna ports 69 Figure 6-16: MSE antenna port#0: Wiener interpolation vs pilot repetition 70 Figure 6-17: MSE antenna port#2: Wiener interpolation vs pilot repetition 70 Page Nguyen Duy Phuong - CB120720 ABBREVIATIONS ACK Acknowledgment ADC Analog to Digital Converter ARQ Automatic Repeat Request CB Code Block CCE Common Channel Element CIR Channel Impulse Response CQI Channel Quality Indicator CP Cyclic Prefix CRC Cyclic Redundancy Check CSI Channel State Information CW Code Word DFT Discrete Fourier Transform DCI Downlink Control Information DL Downlink DMRS Demodulation Reference Signal ENB eNodeB FDD Frequency Division Duplex FEC Forward Error Correction FFT Fast Fourier Transform GI Guard Interval HARQ Hybrid Automatic Repeat Request IDFT Inverse Discrete Fourier Transform IFFT Inverse Fast Fourier Transform ISI Inter-Symbol Interference LLR Log Likelihood Ratio LS Least Square Page Nguyen Duy Phuong - CB120720 LTE Long Term Evolution MAC Medium Access Control MIMO Multiple Input Multiple Output MLSE Mean Least Square Error MMSE Minimum Mean Square Error MRC Maximum Ration Combiner NACK Negative Acknowledgement OFDM Orthogonal Frequency Division Multiplexing PBCH Physical Broadcasting Channel PCCC Parallel Concatenated Convolution Code PCFICH Physical Control Format Indication Channel PDU Packet Data Unit PDSCH Physical Downlink Shared Channel PDCCH Physical Downlink Control Channel PHY Physical Layer PHICH Physical Harq Indication Channel PMI Precoding Matrix Indicator PRACH Physical Random Access Channel PRS Pseudo Random Sequence PUSCH Physical Uplink Shared Channel PUCCH Physical Uplink Control Channel QAM Quadrature Amplitude Modulation QPSK Quadrature Phase Shift Keying REG Resource Element Group RI Rank Indicator RNTI Radio Network Temporary Indicator SC-FDMA Single Carrier Frequency Division Multiplexing SDF Spatial Diversity Frequency SDM Spatial Diversity Multiplexing SISO Single Input Single Output Page Nguyen Duy Phuong - CB120720 SR Scheduling Request SRS Sounding Reference Signal TB Transport Block TDD Time Division Duplex TTI Transmission Time Interval UCI Uplink Control Channel UE User Equipment UL Uplink ZF Zero Forcing Page Nguyen Duy Phuong - CB120720 ABSTRACT In corresponding to ever increasing demands for high quality multimedia services among user’s expectations, 3GPP LTE has been recently considered as the evolution of the UMTS Since downlink is increasingly important contribution in coverage and capacity aspects, special attention has been given in selecting technologies for LTE downlink Novel technologies such as orthogonal frequency division multiplexing (OFDM) and multiple input multiple output (MIMO) tend to be enhance the performance of the current wireless communication systems Utilizing advantages of these two technologies might generate the high data rates and the high capacity Therefore, they have been wisely chosen for LTE downlink Pilot assisted channel estimation is a method in which as signals, called pilots, are transmitted along with data to obtain channel knowledge for proper decoding of received signals This thesis aims at channel estimation for LTE downlink Channel estimation algorithms such as Least Squares (LS), Minimum Mean Square Error (MMSE) haven been evaluated for different channel modes in LTE downlink Performance of these algorithms has been measured in terms of Bit Error Rate (BER) and Symbol Error Rate (SER) Page Nguyen Duy Phuong - CB120720 CHAPTER I : INTRODUCTION TO LTE 1.1 The Emergence of LTE Over the last decade, the significant strides have been generally created in the field of wireless communications The rapid growth of the wireless technology is based on several major contributing factors such as: The developments in RF circuit fabrication and design advanced digital signal processing capabilities and many miniaturization technologies These developments have paved the way to deploy, and more importantly, to deliver reliable and quality communication services In the present times, the indispensable need for wireless data applications has fostered a continuous and renewing interest in the ever so fascinating realm of wireless communications The technology is at a constant evolve challenged to meet the demands of high data rates with tether less quality of service supported with utmost reliability The broadband access to information was one of the several significant milestones in the process of an ever changing technology Broadband provided the means to enhance the usage of wireless applications and has indeed greatly contributed and influenced the quality of life led by people worldwide Faster web-surfing, quicker downloading and real time audio and video streaming have definitely made their immense impact felt globally, thereby contributing in a significant way to the technological market worldwide The goal to maximize the capabilities of broadband was the next challenge, and providing efficient wireless broadband to mobile users was the immediate target The evolution of mobile broadband started initially with the First Generation Cellular Systems, followed by 2G digital cellular systems (GSM, CDMA (IS-95)) and 3G Broadband wireless systems The Long-Term Evolution (LTE) lent itself as a suitable and efficient candidate for delivering mobile broadband LTE was developed by the Generation Partnership Project (3GPP) as an evolution to the still deployed GSM/UMTS networks The spiral surge in the number of fixed-line broadband users ushered the need to develop a mobile Page Nguyen Duy Phuong - CB120720 broadband system that could match the performance and services offered by Digital Subscriber Line (DSL) broadband Focused on this objective, the Radio Access network (RAN) team of the 3GPP organization began to work on the LTE project and the Systems Aspects group initiated work on the System Architecture Evolution (SAE) project The initial study undertaken by these two 3GPP groups was completed by mid-2006 thereby providing a platform for the further development into a standard A new radio access network called Enhanced UTRAN (E-UTRAN), an evolution of UMTS RAN was developed by the LTE group The SAE group developed a new complete IP packet core network architecture called the Evolved Packet Core (EPC) E-UTRAN and EPC combined together was termed as Evolved Packet System (EPS) [1] Goals for LTE: With the aim of outperforming the former technologies, 3GPP had set goals needed to be achieved by LTE The goals were based on achieving better system performance with respect to higher data rates, low implementation costs, reduced system complexity and enhanced overall system throughput A detailed description of is mentioned below [1]: • Performance equivalent with wired broadband: One of the main targets of LTE was to make mobile internet usage much better than residential wired broadband The two vital network performance parameters that propel user experience are high throughput and low latency With the aim of striving towards higher data rates, the 3GPP set the peak data rates for the downlink and uplink at 100 Mbps and 50 Mbps respectively In order to achieve better performance than HSPA, the LTE design target was to achieve a 3-4 times improvement in the average downlink throughput and 2-3 times improvement in the uplink LTE also emphasizes the need for increased cell edge throughput on existing site locations Delay sensitive applications were also considered where the aim was to reduce the round-trip delay time The target round trip latency for LTE radio network is set less Page 10 Nguyen Duy Phuong - CB120720 In all the simulations carried out, the MSE is used as the index to evaluate the channel performance The MSE is given by: HMSE = E|Hideal − Hˆ |2 (6.1) Hideal is a vector comprising of frequency response characteristic of the channel without noise Hˆ is the noisy channel estimates post Wiener filtering in frequency The MSE computation in the frequency domain is done on an OFDM symbol basis However, in the time domain the MSE performance is analyzed on a subframe basis 6.2 Simulation for the pilot channel estimates The frequency response characteristics of the different LTE channel models are plotted Figure 6.1 to Figure 6.3 show the frequency response characteristic |H(f)| of an ideal (noiseless) channel and the channel estimates obtained by the least squares method The noisy channel estimates is extracted according to equation 3.28 The curves give an indication of the shape and closeness of the channel estimates to the ideal channel frequency response It can be observed that for the EPA channel, the frequency response characteristic varies smoothly for the different subcarriers The EVA channel response characteristic has a slightly faster variation compared to EPA while the ETU channel response has a comparatively rapid fluctuation of the channel The frequency selectivity of the channel response is related to the delay spread of the channel model, and hence this translates to the variation in the channel response characteristics The plots shown are the snap shots of one realization of the channel response corresponding to one pilot OFDM symbol of one transmit antenna port A Doppler frequency of Hz and noise variance of 0.5 (linear power) were the considered parameter settings Page 61 Nguyen Duy Phuong - CB120720 Figure 6-1: EPA channel: Plot of ideal channel vs estimated channel Figure 6-2: EVA channel: Plot of ideal channel vs estimated channel Figure 6-3: ETU channel: Plot of ideal channel vs estimated channel Page 62 Nguyen Duy Phuong - CB120720 6.3 Simulation for Wiener interpolation in frequency Figure 6-4: Comparison of ideal channel vs Wiener interpolated channel The noisy channel estimates obtained by method of Least Squares is used for filtering and interpolation of the intermediate subcarriers of each pilot OFDM symbol The ideal channel (noiseless) frequency responses versus the noisy Wiener filtered estimates are plotted Figure 6.4 shows the channel frequency response simulated for an EPA channel and an SNR of 10dB It can be observed that the Wiener interpolated channel estimates are very close to the ideal channel This indicates the goodness of the Wiener filtering in the frequency direction The reliability of the estimates depend on the SNR used, and therefore better channel estimates can be obtained at higher SNR 6.4 MSE performance of LTE channel models Figure 6-5: Wiener interpolation: MSE comparison for LTE channel models Page 63 Nguyen Duy Phuong - CB120720 The performance index for the Wiener filter is based on MSE and is according to Figure 6.5 shows the MSE channel performance for the LTE channel models For this simulation, the Wiener coefficients were computed based on the exponential PDP and rms value of delay spread It is seen that the EPA channel model has the best performance among the three channel models The frequency selectivity of the EPA channel model is minimum thereby translating to a better channel performance 6.5 MSE performance for different power delay profiles Figure 6-6: Performance comparison based on EPA channel Figure 6-7: Performance comparison based on EVA channel Page 64 Nguyen Duy Phuong - CB120720 Figure 6-8: Performance comparison based on ETU channel Figures 6.6 to 6.8 show the performance comparison between and exponential and rectangular power delay profiles In each case the channel is set to EPA, EVA or ETU To provide clarity in understanding the MSE performance implementation, the MSE criterion governed by 6.1 is mentioned again HMSE = E|Hideal − Hˆ |2 (6.2) Hideal is a vector of the ideal frequency channel response for either of EPA, EVA or ETU channel model Hˆ is the filtered and interpolated channel estimates calculated for the exponential and rectangular power spectrum respectively It is worthy to observe the performance trend for the different channel models: It can be seen that for each channel type, the MSE based on the exponential PDP has a higher performance than the rectangular PDP For the EPA channel model the MSE channel performance for the two PDP’s are wide apart For the EVA and ETU channels the difference in performance between the two PDP’s decreases It can be observed that for the ETU channel model, the MSE performances are very comparable in the medium to high SNR range In the case of exponential PDP, the rms value of delay spread is used to generate the Wiener taps while the maximum delay spread value is used for the  rectangular PDP The radio of ( max ) is maximum for the EPA channel i.e.~9 and  rms minimum for the ETU channel i.e.~5 The ratio is computed from table values in Page 65 Nguyen Duy Phuong - CB120720 2.5 The delay spread ratio could be to be a factor that influences the channel MSE performance 6.6 MSE performance for estimated delay spread In a real world scenario the delay spread is not known and it is needed to be estimated The channel correlation functions for different PDP’s are dependent on the delay spread This is evident from equations 3.20 and 3.21 This implies that the Wiener filter coefficients needed for interpolation are generated based on the estimated delay spread values The Wiener performance between the actual delay spread and estimated delay spread is compared Figure 6-9: EPA channel: actual vs estimated delay spread Figure 6-10: EVA channel: actual vs estimated delay spread Page 66 Nguyen Duy Phuong - CB120720 Figure 6-11: ETU channel: actual vs estimated delay spread To compute the MSE two components are considered for each case: The frequency response of the ideal (noiseless channel) with known delay spread The second is the frequency interpolated Wiener estimates with known value of delay spread (indicated in blue) in the plot The frequency response of ideal (noiseless channel) with known delay spread The second component is the frequency interpolated Wiener estimates with estimated value of delay spread (indicated in red) in the plot Figure 6.10 shows the channel performance considering the estimated value of delay spread (indicated in red) for the EPA channel It is seen that the channel performance based on the estimated delay spread is degraded The channel performances are comparable is the low SNR region but widen apart at high SNR’s Figures 6.10 and 6.11 shows that the channel performance is also degraded compared to the ideal case for the EVA and ETU channel In general, the model mismatch between the LTE channel model and exponential model can be a reason for the degraded channel performance 6.7 Doppler spread performance The Doppler spread is estimated based on the cross-correlation of the channel estimates in frequency Specifically, a cross-correlation between pilot OFDM symbols having lags of and are used An MSE measure between the calculated correlations and the actual correlations is computed The Doppler Page 67 Nguyen Duy Phuong - CB120720 frequency corresponding to the minimum MSE is the estimated Doppler spread MSE performance based on estimated Doppler spread: Figure 6-12: MSE performance between actual and estimated Doppler spread Figure 6-13: MSE performance between actual and estimated Doppler spread Figures 6.12 and 6.13 show simulation results for MSE considering actual Doppler’s of 100Hz and 300Hz for an EPA channel model and exponential PDP The Wiener filter coefficients are generated based on the estimated Doppler spread For both cases the channel MSE based on estimated Doppler performance is close to the ideal case (known Doppler) This is evident by the closeness of the curves in Page 68 Nguyen Duy Phuong - CB120720 Figure 6.12 and Figure 6.13 Figure 6-14: SNR comparison: actual vs estimated SNR Figure 6.14 shows the plot for the SNR estimation The simulation is performed for different Doppler frequencies It is observed that for a given Doppler frequency setting, the estimated SNR is fairly accurate at lower SNR’s values The SNR estimation performance drops with increase in Doppler frequency 6.8 MSE performance based on Wiener time interpolation Figure 6-15: Comparison of Wiener time filtering for different antenna ports Figure 6.15 shows the MSE channel performance for the antenna ports having different number of pilot OFDM symbols per subframe The exponential PDP was Page 69 Nguyen Duy Phuong - CB120720 used to generate the filter coefficients In the simulation the delay and Doppler spread was assumed to be known The MSE channel performance between antenna port #0 (4 pilot OFDM symbols per subframe) and antenna port #2 (2 pilot OFDM symbols per subframe) is shown It is seen that for an entire subframe of data being transmitted, antenna port #0 achieves a better performance than antenna port #2 This is due to twice the number of pilot OFDM symbols in antenna port #0, hence contributing to more reliable channel estimates Figure 6-16: MSE antenna port#0: Wiener interpolation vs pilot repetition Figure 6-17: MSE antenna port#2: Wiener interpolation vs pilot repetition The Figures 6.16 and 6.17 shows the MSE performance between the Wiener interpolation method and the Wiener based pilot repetition method The pilot Page 70 Nguyen Duy Phuong - CB120720 repetition is done on a slot basis For an antenna port having pilot OFDM symbols per slot (antenna port#0 and antenna port #1), the average value of the channel estimate per slot is used and repeated for all symbols in that slot In antenna port#2 and antenna port#3, there is only one pilot OFDM symbol per slot and hence one channel estimate is repeated for all the symbols in each slot For the antenna port#0 shown in Figure 6.16 the channel performances are very close to each other at high SNR’s However, the antenna port#2 channel performances shown in Figure 6.17 are much farther apart Page 71 Nguyen Duy Phuong - CB120720 CHAPTER VII: CONSLUTION AND FUTURE WORK In this thesis work, the channel estimation for a MIMO-OFDM system is implemented As the purpose of the thesis was exclusively channel estimation, the gain in using several transmit and receive antennas was limited The MIMO antennas contributed in bringing about an averaging effect in the estimation of the channel parameters such as delay, Doppler and SNR The MSE based 2×1D Wiener filtering is implemented in the channel estimation Simulations were performed to evaluate and compare the MSE performance for the different LTE channel models and power delay profiles Based on the simulation results it can be concluded that the channel performance in frequency direction is largely influenced by the delay spread and the noise power of the channel To evaluate the channel performance of different antenna ports, the MSE of the Wiener filtered (time direction) outputs of the antenna ports were compared As expected, the antenna ports p  {0,1} with more number of pilots OFDM symbols than antenna ports p  {2,3} The next step would be to implement fully compliant LTE system for testing the result The reliability of the algorithms is determined by passing the test cases specified by the 3GPP TS.36.101 Setting up of the test environment, generating test vectors and running the test cases according to the standard 3GPP TS.36.101 will be the work to be done Page 72 Nguyen Duy Phuong - CB120720 REFERENCE [1] Arunabha Ghosh, Jun Zhang, Jeffrey G Andrews and Rias Muhamed, “Fundamentals of LTE”, Pearson Education, 2010.Bob Lantz, Brandon Heller, Nick McKeown, (2010), “A Network in a Laptop: Rapid Prototyping for Software-Defined Networks” [2] “3GPP TS 36.211 V8.7.0”, 3rd Generation Partnership Project; Technical Specification Group Radio Access Network; Evolved Universal Terrestrial Radio Access (E-UTRA); Physical Channels and Modulation [3] “3GPP TS 36.2101 V9.0.0”, 3rd Generation Partnership Project; Technical Specification Group Radio Access Network; Evolved Universal Terrestrial Radio Access (E-UTRA); User Equipment (UE) radio transmission and reception, page 120-125 [4] Huseyin Arslan and Tevfik Ycek “Delay Spread Estimation for Wireless Communications Systems”, 2003 [5] Huseyin Arslan, Leonid Krasny, David Koilpillai and Sandeep Chennakeshu “Doppler Spread Estimation for Wireless Mobile Radio Systems”, 2000 [6] Henrik Schulze and Christian Luders.Theory and Applications of OFDM and CDMA, 1995, page 196-205 [7] Yong Soo Cho, Jaekwon Kim, Won Young Yang and Chung G Kang “MIMO-OFDM Wireless Communications with MATLAB” [8] P.Hoher, S.Kaiser, P.Robertson “Two-dimensional pilot-symbol-aided channel estimation by Wiener filtering In Proceedings IEEE Conference on Acoustics, Speech and Signal Processing (ICASSP’ 97), Munich, Germany (September 1997) [9] P.Hoher, S.Kaiser, P.Robertson “Pilot-symbol-aided channel estimation in time and frequency” in Proc Sixth Communication Theory Mini-Conference in conjunction with IEEE GLOBECOM 97, Phoenix, Arizona, page 90-96, November 1997 Page 73 Nguyen Duy Phuong - CB120720 10 [10] Andrea Goldsmith, Syed Ali Jafar, Nihar Jindal and Sriram Vishwanath “Fundamental Capacity in MIMO channels”, November 8, 2002 11 [11] Xiaohu You, Dongming Wang, Pengcheng Zhu and Bin Sheng “Cell Edge Performance of Cellular Mobile Systems”, June 2011 12 [12] H.Chamkia, A.Omri and R.Bouallegue “Improvement of LTE System Performances by Using a New Pilot Structure” 13 [13] Huseyin Arslan and Sharath Reddy “Noise Power and SNR Estimation for OFDM Based Wireless Communications Systems” Page 74 Nguyen Duy Phuong - CB120720 Page 75 [...]... RADIO CHANNEL AND CHANNEL ESTIMATION 3.1 Characteristics of a Mobile Radio Channel In a mobile radio channel, the transmitted signal propagates in different directions This is due to phenomena such as scattering, diffraction, reflection and shadowing This leads to different copies of the same signal being propagated in multiple directions This is called as multipath propagation A typical mobile radio... the most distinct features of LTE The various multi antenna techniques supported by LTE are: [1] • Transmit diversity: This technique is used to combat the multipath fading in the wireless channel The basic idea is to send replicas(same copies)of the original signal, but coded differently over multiple transmit antennas The transmit Page 35 Nguyen Duy Phuong - CB120720 diversity scheme in LTE is based... CB120720 1.1.1 Choice of technologies for LTE In order to meet the desired service and performance requirements, the design aspects in LTE incorporates vital enabling radio and core network technologies such as OFDM and MIMO One of the major differences between the existing 3G systems and LTE is the underlying modulation scheme used 3G systems such as UMTS and CDMA 2000 are based on Code Division Multiple... signal travel in multiple directions and each of the multipath components are attenuated differently in amplitude and phase depending on the channel conditions experienced by each path As a result, at the receiver different copies of the same signal arrive at different instants of time causing ISI Thus delay spread is the measure of the richness of the multipath channel in a mobile radio environment Delay... within the transmitted symbol time and hence the fading is termed as fast fading Fast fading condition is given as [7] : Ts > Tc and Bs < Bd (3.3) A channel is said to be frequency selective when the amplitude of the channel frequency response varies with frequency and hence frequency selectivity is the effect of fast fading 3.2 Characterization of Fading Channels A complex baseband signal s(t) modulated... Phuong - CB120720 2 CHAPTER II : OVERVIEW OF LTE PHY LAYER The 3GPP defines the LTE radio interface such that the constituent components of the network structure could be evolved independently The network structure comprises of radio access network (RAN) and a core network (CN) The LTE project of 3GPP focused on enhancing and optimizing 3GPP’s overall radio access architecture, while the Evolved Packet... M sq ) is the number of modulation symbols in each codeword and depends on the modulation scheme QPSK, 16QAM, 64QAM and BPSK are the different modulation schemes supported by LTE for the different physical channels • Layer mapping and Precoding: Layer mapping and precoding define the processing needed for MIMO transmission and reception The overall processing involves the mapping of the incoming codewords... component are attenuated to different extent At the receiver, each multipath component arrives at different instants with varying signal strength This effect is called fading and is one of the effects of multipath propagation The extent of fading depends on the frequency selectivity of the channel and rate of time variation of the channel.[7] Figure 3-1: A typical mobile radio environment When the coherence... vary with time Noise is the distorting factor to be considered in the analysis In a static channel model the channel performance measurement is carried out in an Additive White Gaussian Noise (AWGN) environment • Multipath fading propagation channel model: In a practical real word scenario the radio channel is time variant The time-varying characteristic of the mobile radio channel causes the transmitted... rectangular PDP 3.3 Orthogonal Frequency Division Multiplexing (OFDM) Orthogonal Frequency Division Multiplexing is a multicarrier modulation scheme successfully used in most of the wireless broadband systems such as Digital Subscriber Lines (DSL), Wireless LAN (802.11a/g/n), Digital Video Broadcasting, and most recently beyond 3G cellular technologies such as WiMAX and LTE The popularity and success in OFDM