CHAPTER-1

INTRODUCTION

Todays communication services have different constraints concerning data rates, latency and reliability. Bandwidth, power, cost and complexity are among the challenges for the designing of a communication system. With MIMO, the system spectral efficiency can be increased by allowing the transmission using several spatially separated data streams. Based on the assumption of ideal propagation environment, the capacity or spectral efficiency of MIMO channels grows approximately linearly with the number of transmitter antennas. IT is widely acknowledged that reliable fixed and mobile wireless transmission of video, data, and speech at high rates will be an important part of future telecommunications systems. One way to get high rates on a scatteringrich wireless channel is to use multiple transmit and/or receive antennas. In, theoretical and experimental evidence demonstrates that channel capacity grows linearly as the number of transmit and receive antennas grow simultaneously. Early uses of multiple transmit antennas in a scattering environment achieve reliability through diversity, where redundant information is sent or received on two or more antennas in the hope that at least one path from the transmitter reaches the receiver. To keep the transmitter and receiver complexity low, linear processing is often preferred. To achieve the high data rates promised in, however, new approaches for spacetime transmission are needed. One such approach is presented in where a practical scheme, called V-BLAST (Vertical Bell Labs Layered SpaceTime), encodes and decodes rates of tens of bits per second per hertz (b/s/Hz) with 8 transmit and 12 receive antennas. The VBLAST architecture breaks the original data stream into sub streams that are
2

transmitted on the individual antennas. The receiver decodes the sub streams using a sequence of null and canceling steps.

MIMO

(multiple-input

multiple-output)

technology

significantly

enhances system performance. The extra degrees of freedom afforded by the multiple antennas can be used for increasing bit rates through spatial multiplexing [1][4] or for improved diversity order through space-time coding techniques [5], [6]. However, multiple antenna deployment requires multiple RF chains (consisting of amplifiers, analog to digital converters, etc.) that are typically very expensive. There is, therefore, considerable incentive for lowcost, low-complexity techniques with the benefits of multiple antennas. Optimal antenna subset selection is one such technique. A selection of antenna elements, which are typically much cheaper than RF chains, is made available at the transmitter and/or receiver. Transmission/reception is performed through the optimal subset.

With space time coding, simultaneous diversity and coding gains can be obtained in addition to spectral efficiency. To increase system reliability, MIMO combined with Bit-interleaved Coded Modulation has been simulated for higher order PSK which performs well in mobile fading channels and BER performance is achieved at very low SNR. The goal of next generation networks (NGN) is to provide higher data rates as compared to the existing networks while granting the same degree of user mobility. NGNs are expected to deliver more advanced features such as enhanced multimedia, smooth streaming video, universal access and portability across all types of devices. These networks are envisioned to have worldwide roaming capability and will actually connect the entire globe and be operable from any location on earth or above the surface of
3

earth. The core network is designed based on IP which tolerate a variety of radio access protocols. Future mobile networks are to provide anywhere, anytime, communications services to users requiring high data rates much greater than current networks. The applications could be high quality wireless video conferencing requiring up to 100 Mbps or wireless virtual reality requiring up to 500 Mbps. Wireless local area networks have higher data rates as compared to wide area networks such as 3G networks. These networks cover small areas often hotspots in shopping malls, airports, hotels or office buildings. The IEEE802.11a based WLAN delivers at about 54 Mbps using the 5-GHz band. The physical layer of this standard is based on orthogonal frequency division multiplexing (OFDM). To achieve the ambitious goal of high data rates for next generation wireless networks with limited power and bandwidth at affordable complexity brings about tremendous challenges. MIMO combined with OFDM is one of the promising techniques to achieve the required data rates for next generation networks.

OFDM (orthogonal frequency division multiplexing) technique has been adopted as the standards in the several high data rate applications, such as Europe DAB/DVB (digital audio and video broadcasting) system, high-rate WLAN (wireless local area networks) such as IEEE802.11x, HIPERLAN II and MMAC (multimedia mobile access communications), and terrestrial DMB (digital multimedia broadcasting) system. OFDM system transmits information data by many sub-carriers, where sub-carriers are orthogonal to each other and sub-channels are overlapped so that the spectrum efficiency may be enhanced. OFDM can be easily implemented by the IFFT (inverse fast Fourier transform) and FFT (fast Fourier Transform) process in digital domain, and has the property of high-speed broadband transmission and robustness to multi-path interference, frequency selective fading. However, OFDM signal has high
4

PAPR (peak to average power ratio) because of the superimposition of multicarrier signals with large number of sub-carriers. The high PAPR makes the signal more sensitive to the nonlinearities of the HPA (high power amplifier) and result in signal distortion when the peak power exceeds the dynamic range of the amplifier. To transmit the high PAPR signal without distortion requires more expensive power amplifier with high linearity and wider dynamic range. Besides, the non-linear distortions due to clipping and amplification effects in the transmitted signal will lead to both in-band and out-of band emissions. The former provokes BER degradation whereas the later results in spectral spreading. Combining the advantages of space-time coding and OFDM is attractive in wireless system designs. This involves coding across space and frequency, which is often referred to as space-frequency coding (SFC).

A way to do space-frequency coding is to take the space time codes (e.g., Alamouti code), and apply them in the frequency dimension instead of time dimension. That is, instead of mounting the space-time coded symbols on multiple time slots, they are mounted on multiple OFDM subcarriers. Use of orthogonal space-time block codes (OSTBCs) in the frequency dimension is attractive for SFC OFDM because of their low complexity decoding (i.e., single symbol decode-ability) and suitability for fast fading channels. A spacefrequency block coded (SFBC) OFDM scheme which uses Alamouti code in the frequency dimension is defined in [8] for high mobility broadband wireless access. For the time dimension OSTBCs to be single symbol decodable, the often made `quasi-static' (QS) assumption (i.e., fade remains constant over one block time, which is valid only in slow fading channels) is essential, the violation of which results in an error-floor. Rapid time variations in the fading process result in such a violation. In SFBC OFDM systems, on the other hand,
5

the QS assumption gets violated in the frequency dimension in highly frequency-selective channels (i.e., different subcarriers, and hence symbols belonging to the same SFBC block mounted on different subcarriers, see different channel gains), even if time-variations in the fading process is very slow. The severity of this effect depends on the channel length L, power delay profile of the channel, and the SFBC block size. In highly frequency selective channels (i.e., large L), this QS assumption violation becomes a source of significant inter-symbol interference (ISI) in the frequency dimension in SFBC OFDM. If left uncared for, this results in error floors.

Further, in any OFDM system, the orthogonality among subcarriers is lost if the channel changes within an OFDM symbol duration, which results in inter-carrier interference (ICI). Thus, in addition to the issue of ISI caused due to frequency selectivity of the channel, SFBC OFDM experiences ICI caused due to time-selectivity of the channel (i.e., channel varying within one OFDM symbol duration) . Like ISI, ICI, if uncared for, also will result in error floors. Attempts have been made in the literature to cancel ICI in MIMO OFDM systems. Proposed ICI-mitigating block linear filters for STBCOFDM. However, they do not consider the loss of QS assumption in large delay spread channels. Improvement in performance is possible in SFBC-OFDM if both ISI (Due to loss of QS assumption) as well as ICI (due to time-selectivity) can be estimated and cancelled. In this paper, we propose a linear parallel interference cancellation (PIC) approach to mitigate the effects of both ISI and ICI in SFBCOFDM. The proposed detector estimates (using soft output values) and cancels the ISI in the first step, and then estimates and cancels the ICI in the second step. This two step procedure is carried out in multiple cancellation stages. We evaluate the performance of the proposed detector for different codes including i) rate- 1 Alamouti code, and ii) rate-2/3 G5 code [12], for varying degrees of
6

time selectivity (different speeds) and frequency selectivity (different channel lengths, L). We show that the proposed detector effectively cancels the ISI and ICI in high mobility, large delay spread channels.

CHAPTER-2

EXISTING SYSTEM

Full

Rate or FR or GSM-FR or GSM

06.10 was

the

first

digital speech

coding standard used in the GSM digital mobile phone system. The bit rate of the codec is 13 Kbit/s, or 1.625 bits/audio sample (often padded out to 33 bytes/20 ms or 13.2 Kbit/s). The quality of the coded speech is quite poor by modern standards, but at the time of development (early 1990s) it was a good compromise between computational complexity and quality, requiring only on the order of a million additions and multiplications per second. The codec is still widely used in networks around the world. Gradually FR will be replaced by Enhanced Full Rate (EFR) and Adaptive Multi-Rate (AMR) standards, which provide much higher speech quality with lower bit rate. Full-rate 22 STC design which attempts to maximize both the diversity gain and the coding gain, while leading to an optimum detector of reduced complexity. More specifically, the proposed STC is a full-rate, full-diversity 22 space-time code whose optimum receiver has a complexity that is only proportional to M2, where M is the size of the signal constellation. Thus, the number of Euclidean distance computations in the optimum detector is reduced to 162 = 256 for a 16QAM signal constellation and to 642 = 4,096 for a 64-QAM signal constellation. Comparing these numbers to those associated to the Golden code (or Matrix C), it becomes clear that this code makes the implementation of fullrate, full-diversity 22 STCs with optimum receiver realistic.

STC DESIGN CRITERIA We start with a brief discussion of the most common design criteria for STCs when the transmitter does not have any channel state information. For the 22 transmission, we write
9

R = HX + Z, Where H is the 22 channel matrix with the entries of complex channel gains, X is the 22 codeword matrix

In (1), R includes the received signals, and Z denotes the matrix of additive noise samples. Code optimization generally relies on the analysis of pair wise error probability (PEP) P(X X) which is the probability that X is detected while X is transmitted. The goal is to minimize the error probability, but the analysis of exact error performance is difficult, and therefore one resorts to the union bound

Where X denotes the codebook of the space-time code. A Chernoff bound analysis of the PEP was performed in and this analysis led to the following design criteria: 1) Rank Criterion: In order to achieve the maximum diversity, the diversity gain,

Where = X X, should be maximized. If H is full Rank for all code word pairs, then the code is said to have Full diversity.

10

2) Determinant Criterion: The coding gain is defined as

Where) ( l denotes the lth largest Eigen value of the enclosed matrix. In order to obtain the best performance, the coding gain should be maximized for a given average transmits power. It should be noted that, for high signal-to-noise ratio (SNR) values, the most important parameter is the diversity gain which dominates the steepness of the bit error rate (BER) curve. Afterwards, it is the coding gain which should be maximized. In the sequel, we will directly optimize the proposed code such that the coding gain is maximized. In light of the above design criteria, this directly ensures that the optimized code will have full diversity. DRAWBACKS: 1. FRC has very high decoding complexity, which grows exponentially with the number of transmitted symbols 2. FRC requires number of iterative process for decoding the transmitted signal in MIMO 3. FRC depends on the Channel condition and noise level LITERATURE SURVEY:

S. M. Alamouti, A Simple Transmit Diversity Technique for Wireless Communications, IEEE J. Sel. Areas Commun. , vol. 16, No. 8, pp. 1451 1458, Oct. 1998.

11

V. Tarokh, N. Seshadri, and A. R. Calderbank, Space-time codes for high data rate wireless communications: Performance criterion and code construction, IEEE Trans. Inf. Theory, vol. 44, no. 2, pp. 744--765, Mar. 1998. H. Yao and G. W. Wornell, achieving the full MIMO diversitymultiplexing frontier with rotation-based space-time codes, in Proc. Allerton Conf. Commun., Control, and Computing, Monticello, IL, Oct. 2003.

12

PROPOSED SYSTEM SPACE-TIME coding is one of the main methods in order to exploit the capacity of multiple-input multiple-output (MIMO) channels [1]. Since STC techniques use both time and spatial domains for coding data symbols, diversity and spatial multiplexing can be combined achieving robustness at the receiver with a higher data rate transmission. As a result, STC techniques have been incorporated in many of the last-generation wireless communications systems, including the new generation of terrestrial and mobile digital video broadcasting (DVB) standards. If STC is joined to multi-carrier modulation, such as orthogonal frequencydivision multiplexing (OFDM), space frequency block coding (SFBC) can be performed. This way, codewords are fed into adjacent carriers of the two consecutive OFDM symbols, translated to the time domain and transmitted through several transmit antennas. This transmission scheme is usually combined with bit-interleaved coded modulation (BICM) giving good diversity results in a wireless communication link [2]. In order to achieve the full MIMO diversity-multiplexing frontier [3], the proposals for the future generations of terrestrial, portable and mobile digital video broadcasting standards, such as DVB-NGH, focus on the combination of both diversity and spatial multiplexing [4], [5] through full-rate full-diversity (FRFD) codes [6]. The main drawback of full-rate codes arises from their very high decoding complexity, which grows exponentially with the number of transmitted symbols per codeword. In order to reduce the complexity of the detection process, hard detection techniques such as sphere decoding (SD) or low complexity STC designs [7], [8] can be used. Nevertheless, when iterative decoders, such as turbo or low-density parity check (LDPC) codes, are included in the reception chain, soft information on the conditional probabilities for all possible transmitted symbols is required in the form of log-likelihood ratios (LLR). Moreover, the computation of the LLRs for
13

the whole set of transmitted symbols is unfeasible, especially for large constellation sizes. Hence, the soft MIMO detector has to select a group of candidates to be fed to the decoder in order to compute the required LLRs. Several algorithms that serve this purpose can be found in the literature, such as list sphere detection (LSD) [9], near-optimal soft SD [10], list fixed-complexity sphere detection (LFSD) [11], QR decomposition associated with the Malgorithm (QRM-MLD) [12] or bounded soft sphere detection (BSSD) [13]. The contribution of this paper is a new low-complexity soft detection algorithm for FRFD SFBC and its assessment in an LDPC-based BICM scenario. The generation of the candidate list is carried out by means of a fixed tree search, whose complexity does not depend on the channel conditions or the noise level. Since the complexity of the proposed soft detector is closely linked to the architecture of the tree search, we analyze different tree configurations in order to find the best balance between complexity and performance. Simulation results are provided for the Golden [14] and the FRFD Sezginer-Sari (SS) codes [8] in a DVB-T2 framework [15]. Although our research has been carried out on a terrestrial TV system, the results can be generalized for any MIMO bit interleaved coded modulation (MIMO-BICM) scheme. The remainder of the paper is organized as follows: Section II details the system model, focusing on soft detection and LLR calculation. The design of the new low-complexity detection algorithm is presented for two FRFD SFBCs in Section III, while Section IV shows the simulation-based performance of the proposed receiver in DVB-T2 scenarios.

14

LDPC:

A Tanner graph is a bipartite graph that describes the parity check matrix H There are two classes of nodes: o Variable-nodes: Correspond to bits of the codeword or equivalently, to columns of the parity check matrix
15

 There are n v-nodes o Check-nodes: Correspond to parity check equations or equivalently, to rows of the parity check matrix There are m=n-k c-nodes o Bipartite means that nodes of the same type cannot be connected (e.g. a c-node cannot be connected to another c-node) The ith check node is connected to the jth variable node iff the (i,j)th element of the parity check matrix is one, i.e. if hij =1 o All of the v-nodes connected to a particular c-node must sum (modulo-2) to zero LOW DENSITY PARITY CHECK CODES: Proposed by Robert G. Gallager (1960) Sparseness of matrix and graph descriptions o Number of 1s in H grows linearly with block length o Number of edges in Tanner graph grows linearly with block length Randomness of construction in: o Placement of 1s in H o Connectivity of variable and check nodes Iterative, message-passing decoder o Simple local decoding at nodes o Iterative exchange of information (message-passing)
16

 Now called regular LDPC codes Parameters (n,j,k) o o n = codeword length j = # of parity-check equations involving each code bit

= degree of each variable node o k = # code bits involved in each parity-check equation

o = degree of each check node Locations of 1s can be chosen randomly, subject to (j,k) constraints. LDPC examples

Regular LDPC Code Tanner Graph

17

Properties of Regular LDPC Codes Design rate: R(j,k) =1 j/k Linear dependencies can increase rate Design rate achieved with high probability as n increases Example: (n,j,k)=(20,3,4) with R = 1 3/4 = 1/4. For j 3, the typical minimum distance of codes in the (j,k) ensemble grows linearly in the codeword length n. Their performance under maximum-likelihood decoding on BSC (p) is at least as good...as the optimum code of a somewhat higher rate. [Gallager, 1960]

18

Performance of Regular LDPC Codes

19

LDPC Codes and Their Applications

Low Density Parity Check (LDPC) codes have superior error performance o 4 dB coding gain over convolution codes Standards and applications o 10 Gigabit Ethernet (10GBASE-T) o Digital Video Broadcasting (DVB-S2, DVB-T2, DVB-C2) o Next-Gen Wired Home Networking (G.hn) o WiMAX (802.16e) o WiFi (802.11n) o Hard disks

20

o Deep-space satellite missions BIT INTERLEAVER: In digital transmission a voice sample is converted to 8 binary bits. These bits will be sent in 3.9us to the other end through the transmission media if the multiplexing used is CEPT (Committee of European Post and Telecom). Is it necessary to send these 8 bits in serial in 3.9 us. Otherwise if we use 8 media in 488ns the sample information will be received at the other end. The latter method is not used due to the following reasons. 1. Instead of one transmission media 8 transmission media to be used. Hence cost is high. 2. Even 8 transmission medias deployed it will not have similar characteristics for the buts in transmission. Leading the receiver to deploy additional techniques to synchronize the bits for a word. Here a word means a sample information of 8 bits. Hence in the basic PCM of 2.048 Mbps always 8 bits pertaining to sample information is sent serially. Since there are 32 Time Slots (30 channels are multiplexed in the CEPT system) . All the bits pertaining to a sample are serially interleaved and sent in one transmission media. But when a multiplexing is carried out in the higher order the basic information to the higher order multiplexer is a bit pertaining to a primary tributary of 488ns and not 8 bits pertaining to a sample. Two interleaving methods can be adopted. They are as follows 1. Bit Interleaving- As explained earlier four bits of 4 primary tributaries are multiplexed together as one time slot on the higher tributary. This is shown as follows. Figure:

21

Word (Byte) Interleaving- In this method the scanning of each bit is similar to above while transmission on the higher order PCM, all these scanned primary information for a particular Time Slot of Primary tributary will be stored and sent serially in the higher order PCM as one time slot but not in 3.9us. In the case of 8.448 Mbps this time slot will carry 8 bits in 117*8 = 936ns. In the practical situation Bit Interleaving deployed due to treatment of practical primary PCMs non alignment of clocks and will be discussed in detail. While the word Interleaving is deployed in circuit switching systems where all the primary PCMs to the switching network is synchronized. SFBC: There are two modes in which MIMO can improve LTE system performance. One mode is spatial multiplexing, in which the vertical Bell 1 Labs layered space-time (V-BLAST) codes [6] are employed to obtain capacity gain. The other mode is transmitting diversity, in which the space-frequency block codes (SFBC) [7] are used to obtain diversity gain. The SFBC used in LTE is shown in Table 1 and Table 2, in which the Alamouti space-time block code (STBC) [8] is used by replacing the time domain with the frequency domain. The resulting SFBC can realize full space diversity but is not guaranteed to achieve
22

full (space and frequency) diversity [9]. As the Alamouti STBC which suffers performance degradation in presence of high Doppler [10], the LTE SFBC cannot achieve optimal performance unless the channel remaining constant over at least two consecutive tones.

It can be seen from Table 1 and Table 2 that the 4-antenna SFBC is a simple superposition of two 2-antenna SFBC. If the channel coding and interleaving are ignored, the raw bit error ratio (BER) performance of the 4-antenna SFBC is identical to the performance of the 2-antenna SFBC. Therefore, in this paper only the 2-antenna SFBC is considered. The rest of the paper is organized as follows. Section 2 introduces the system model for theoretical analysis and the assumption of channels. Section 3 theoretically analyzes the average signal to interference-plus-noise ratio (SINR) with imperfect channel estimation, delay spread, and correlated channels. Section 4 verifies the theoretical analysis by numerical simulation as well as provides the simulation results with ITU pedestrian and vehicular channel models [11]. Section 5 comes to concluding remarks. 2 System Model
23

2.1 SFBC and decoding We use the following notation: implies that z is a complex

circularly symmetric Gaussian random variable with mean _ and variance _2; in is n n unity matrix. For simplicity, we consider there are Nt = 2 antennas at the transmitter and Nr = 1 antenna at the receiver. An SFBC symbol is transmitted from both antennas at two consecutive OFDM tones.

The received symbol is given by

where n0, n1 i.i.d N (0; _2n ) are the additive noises. In (2), hij , i; j {0; 1}, denotes the channel response at the OFDM tone i from transmit antenna j. Each tone is assumed to experience flat Rayleigh fading, i.e., hij N (0; 1), i; j {0; 1}. It should be noticed that h0j and h1j , j {0; 1}, may not be identical because of the delay spread. Besides, hi0 and hi1, i {0; 1}, may not be independent because of the correlation between antennas. Assuming that the receiver estimate channel by pilot symbols, the estimated channel responses can be modeled by [12]

24

where E is a random matrix with each element eij i.i.d N (0; _2 e ), i; j {0; 1}. The desired signal after maximum likelihood decoding [13] is given by

Eq. (6) is the desired components, (7) lead to self-interference, and (8) are additive noises. SFBC WITH OFDM: Use of space-frequency block coded (SFBC) OFDM signals is advantageous in high-mobility broadband wireless access, where the channel is highly time- as well as frequency-selective because of which the receiver experiences both inter-symbol interference (ISI) as well as inter-carrier interference (ICI). ISI occurs due to the violation of the `quasi-static' fading assumption caused due to frequency- and/or time-selectivity of the channel. In addition, ICI occurs due to time-selectivity of the channel which results in loss of orthogonality among the
25

subcarriers. In this paper, we are concerned with the detection of SFBC-OFDM signals on time- and frequency-selective MIMO channels. Specially, we propose and evaluate the performance of an interference cancelling receiver for SFBC-OFDM which effectively alleviates the effects of ISI and ICI.

26