Phase Noise Detection

Figure 4.8 shows the performance without the proposed phase noise detection and the profile of PER = 1. Figure 4.9 is the same condition except with our design. Compare these two figures, we can find that the phase noise tolerance is surprisingly improved.

Note that the phase fluctuates dramatically over boundary (±π) when loop bandwidth is larger than 6 ppm, so performance degrades fast. Another thing deserve to be mentioned is that the detection algorithm doesn’t work when loop time constant is less than 7 kHz when packet length equals to 1000 since it needs two peaks. Although the tolerance of our method is not enough large, the serious condition won’t happens or no one can handle that situation. The so called cycle slip occurs when the carrier frequency changes from one edge to another edge quickly, e.g. from -50 ppm to 50 ppm with 100 kHz. In this situation, even we have good algorithm can still not save the performance loss.

Table 4.4 Comparison of the tolerance without and with phase noise detection.

Estimation property

Figure 4.8 Performance without phase noise detector and the profile when PER = 1.

Figure 4.9 Performance with phase noise detector and the profile when PER = 1.

Chapter 5

THE PROPOSED ARCHITECTURE

5.1 Architecture of Adaptive Equalization

The whole architecture of the proposed adaptive equalization can be divided into three parts, the CFR error tracking, the property measurement and the others. The architecture of the algorithm is depicted in Figure 5.1. At first, distorted long preamble is divided the known training symbols in advance. We obtain the initial CFR and put it into the 5-tap smooth filter. The smoothed CFR will be used for data compensation. The CFR error tracking loop contains five components. The upper adder and multiplier are used to calculate normalized de-mapping error vector. Then the results representing the mean of normalized constellation error vectors will be stored in the accumulator. After multiplying the smoothed CFR, the residual estimation error is induced. The most important part of this architecture is the property measurement which consists of two pairs of MSE calculator and a comparator. The MSE calculator is composed of one adder, square device, four shift registers and the summation block. As implied by the name, it computer the MSE of the equalized pilots and the desired pilots. In the end, the comparator will decide if we should update the CFR, since this output of the property measurement controls the multiplexer for equalization. The hardware cost listed in Table 5.1 includes four multipliers, four adders and about 162 kilo-bytes memory space.

Figure 5.1 The hardware architecture of adaptive equalization.

Table 5.1 Hardware complexity of the proposed adaptive equalization.

Multiplier Adder Register (B)

Quantity 4 6 1.6 k

5.2 Architecture of Phase Noise Detection

Figure 5.2 shows the architecture of the phase noise detection. In the beginning, the residual frequency offset, C0,l, and the corresponding OFDM symbol number, l, are delivered form PR. We use the remaining frequency offset to calculate the tendency of the wave came of phase noise as described in section 3.3. The search window is implemented with ten 1-bit shifter registers. The half of the registers is added then compared with the error threshold for search peak. Here the threshold is set to 3 over 5, which mean we can declare the peak is found if there are at least three in five taps.

As soon as peak is found, the corresponding symbol number is passed to the table for estimating loop time constant, Δω. Then we use the result for estimating another parameter. The difference of the remaining frequency offset, their corresponding symbol numbers and the estimated loop time constant are the input of the table containing formula for computing loop bandwidth. The comparator of upper one is used to control the switch of tracking mechanism. Another comparator is used to avoid the loop bandwidth out of bound and will make B dispersing. The hardware cost of the proposed phase noise detector listed in Table 5.2 includes four adders and 109 bytes memory space in rough. The compensation architecture here is ignored because it only needs to fix the NCO table for adding sine calculation.

Figure 5.2 The hardware architecture of phase noise detection.

Table 5.2 Hardware complexity of the proposed phase noise detection.

Multiplier Adder Register (B)

Quantity 0 4 109

Chapter 6

CONCLUSION AND FUTURE WORK

6.1 Conclusion

This thesis proposes a novel scheme for channel equalization in OFDM receivers. The proposed algorithm can use de-mapping results and pilots to adjusting estimate channel frequency response in singular multipath environments. From simulation results, the average estimation error of the proposed algorithm is small enough for little system performance loss under different multipath channel. And the convergence speed is much faster at the same time. Besides, the proposed method is suitable for implementation issue compared with the reference designs.

By the way, a phase noise with CFO model is also constructed in this thesis. It is found that the joint effects of phase noise and CFO degrade the system performance dramatically. Thus we propose a novel estimation algorithm and a robust compensation scheme which is shown to work well under loop bandwidth 11 ppm, loop time constant 25 kHz, and maximum CFO tolerance 50 ppm at 2.4 GHz carrier frequency. All of the estimations are done in the known pilots.

Therefore, the proposed algorithm can enhance the performance of OFDM systems and is possible to achieve both small and low-cost systems.

6.2 Future Work

In the future, high QAM constellation which will be more sensitive to non-ideal effects such as CFO and phase noise may be used for higher data rate. Multi-input

multi-output OFDM (MIMO-OFDM) systems are also gaining prominence in high data rate applications. However, the above problem in this thesis will limit the system performance. That means a more robust scheme has to be applied to overcome these impairments. Although the CFO estimation mechanism is good enough, it still needs some adjustment when IQ mismatch occurs [8]. Besides, computation in low-SNR points for CFO estimation can be eliminated for high performance and low complexity, i.e. coarse tune using two autocorrelations with 4 point per short symbol and fine tune using one autocorrelation with 32 points per long symbol [10].So some extensions to the research presented in this thesis will be included in our future work.

A. In the future, the work is to improve the detection range of the proposed method. Additionally, we will need some mechanism for special case in the proposed method.

B. This thesis proposed the phase noise estimation in single-input single-output OFDM (SISO-OFDM) system. In the future, we will try to derive a model for the effect of phase noise and CFO in MIMO-OFDM system and develop a compensation technique for the impairment.

References

[1] IEEE 802.11a, IEEE Standard for Wireless LAN Medium Access Control and Physical Layer Specifications, Dec. 1999.

[2] IEEE 802.11b, IEEE Standard for Wireless LAN Medium Access Control and Physical Layer Specifications, Sept. 1999.

[3] IEEE 802.11g, IEEE Standard for Wireless LAN Medium Access Control and Physical Layer Specifications, June 2003.

[4] Hung-Kuo Wei, A Frequency estimation and compensation method for high speed OFDM-based WLAN system, National Chiao Tung University, Taiwan, June 2003, Master’s thesis.

[5] Takeshi Onizawa, Masato Mizoguchi, Tetsu Sakata, Masahiro Morikura, “A simple adaptive channel estimation scheme for OFDM systems,” in Proc. IEEE Vehicular Technology Conference (VTC 1999-Fall), vol. 1, pp. 19-22 Sept. 1999.

[6] Sun-Wook Kim, Kyun-Hyon Tchah, “Performance analysis of adaptive equalizer design for OFDM wireless LAN,” IEEE Transactions on Consumer Electronics, vol. 50, pp. 512-516, May 2004.

[7] Yi-Hsin Yu, A Channel Equalizer for OFDM-based Wireless Access Systems, National Chiao Tung University, Taiwan, June 2004, Master’s thesis.

[8] Ming-Fu Sun, Joint Estimation and Compensation of I/Q Mismatch and Carrier Frequency Offset in OFDM Systems, National Chiao Tung University, Taiwan, June 2005, Master’s thesis.

[9] Chien-Jen Hung, A Differential Decoding Based Baseband Processor for DSSS Wireless LAN Applications, National Chiao Tung University, Taiwan, June 2003,

Mater’s thesis.

[10] Lin-Hung Chen, A Low Complexity Frequency Synchronizer for OFDM-based Wireless Access Applications, National Chiao Tung University, Taiwan, June 2005, Mater’s thesis.

[11] John Terry and Juha Heiskala, OFDM Wireless LANs: A Theoretical and Practical Guide. Sams, Indianapolis, Indiana, 2002, ISBN 0-672-32157-2.

[12] Zhiguo Zhang, Shing-Chow Chan, Hui Cheng, “Robust Adaptive Channel Estimation of OFDM Systems in Time-Varying Narrowband Interference,” in Proc. IEEE International Symposium on Circuits and Systems (ISCAS 2005), pp.

1722-1725, May 2005.

[13] Xuan Huan Nguyen, Jinho Choi, “Reduced-rank adaptive time-domain channel estimation for OFDM systems,” in Proc. IEEE International Conference on Communications Systems (ICCS 2004), pp.135-139, Sept. 2004.

[14] Taewoo Han, Xiaohua Li, “Blind adaptive equalization of OFDM transmission with insufficient cyclic prefix,” in Proc. IEEE International Conference on Modern Problems of Radio Engineering, Telecommunications and Computer Science (TCSET 2004), pp. 213-216, Feb. 2004.

[15] Hewavithana, T.C., Brookes, D.A., “Blind adaptive channel equalization for OFDM using the cyclic prefix data,” in Proc. IEEE Global Telecommunications Conference (GLOBECOM'04), vol.4, pp. 2376-2380, Nov./Dec. 2004.

[16] Fan Yang, Wee Ser, “Adaptive semi-blind channel estimation for OFDM systems”, in Proc. IEEE Vehicular Technology Conference (VTC 2004-Spring), vol.3, pp. 1773-1776, May 2004.

[17] Doukopoulos, X.G., Moustakides, G.V., “Adaptive algorithms for blind channel estimation in OFDM systems”, in Proc. IEEE International Conference on Communications (ICC 2004), vol.4, pp. 2377-2381, June 2004.

[18] K. Fazel, S. Kaiser, Multi-Carrier and Spread Spectrum Systems. Wiley, 2003, ISBN 0-470-84899-5.

[19] Moose, P.H., “A technique for orthogonal frequency division multiplexing frequency offset correction,” IEEE Transactions on Communications, vol. 42, pp. 2908-2914, Oct. 1994.

[20] Shynk, J.J, “Frequency-domain and Multirate adaptive filtering”, IEEE Signal Processing Magazine, vol. 9, pp. 14-37, Jan. 1992.

[21] Eur. Telecommun. Stand., Radio Broadcasting Systems: Digital Audio Broadcasting to Mobile, Portable and Fixed Receivers, Feb.1995.

[22] Eur. Telecommun. Stand., Digital Vedio Broadcasting: Framing Structure, Channel Coding, and Modulation for Digital Terrestrial Television, Aug. 1997.

[23] Garcia Armada, A, “Understanding the effects of phase noise in orthogonal frequency division multiplexing (OFDM),” IEEE Transactions on Broadcasting, vol. 47, pp.135–159, June 2001.